I’ve had the privilege of knowing Eugene for nearly 20 years. We were both platoon leaders and company XOs in 1st Tank (1-72 AR) from 2007-2008 in Camp Casey, Korea. Our paths continued to Fort Lewis, where we deployed together to Afghanistan with 5/2 SBCT. We’ve kept in touch over the years and it’s been great to see his career progress and family thrive.

We are honored to have him join Exia Labs for his Skillbridge. Eugene brings a wealth of contracting and program management expertise from both the U.S. Army and the Intelligence Community.

Eugene is an award winning Acquisition Officer who graduated from the United States Military Academy and the Naval Postgraduate School. He is an Airborne Ranger with combat deployments to Afghanistan and Iraq along with many other missions all throughout Africa, Asia, and Europe. He commanded Abrams, Strykers, and Bradleys as a company grade officer and as a field grade transferred to Army Acquisitions and conducted contingency contracting operations for high priority operations worth billions of dollars such as Operation Inherent Resolve, Operation Allies Welcome and the European Assure/Deter/Reinforce mission.

During his recent tenure at the National Geospatial-Intelligence Agency, he served in the International Technologies Office as a Program Manager where he forged vital partnerships with various Department of War research labs and industry partners, seamlessly integrating their cutting-edge technologies with NGA’s enterprise systems. In addition to his technical accomplishments, Eugene contributed to advancing cybersecurity and AI initiatives at NGA.

He has done it all before joining Exia for his Skillbridge Fellowship!

Mark Beall, former senior Defense Department official: “Claude is seen as the best model on the market in the military.”

Nicholas Chaillan, Founder Ask Sage: “I deployed Claude across the DoD on classified networks. It is one of the best models available to our warfighters.”

The "beltway elite" have long maintained that Anthropic’s Claude is the frontier model for defense. But it has been banned across the government. Anthropic is challenging that decision with a lawsuit. Meanwhile, OpenAI has aggressively stepped into the fold and secured a contract with the Department of War.

If you follow the headlines, it sounds like these models are performing Ender’s Game-level analysis. The truth is much more mediocre. Even on simple military tasks—like basic terrain analysis—the world's most powerful frontier models are struggling to provide actionable intelligence.

The Test: 8 Square Miles of Kharg Island

Kharg Island is arguably the most strategically sensitive piece of land in Iran. It handles approximately 90% of Iran’s crude oil exports. Kharg is only 8 square miles. As of this week, the 82nd Airborne Division has been ordered to deploy to the Middle East and the 11th and 31st Marine Expeditionary Units (embarked on the USS Tripoli and USS Boxer) are closing in on the Persian Gulf.

You would expect the world’s most powerful frontier AI models to be able to analyze a piece of terrain smaller than Mercer Island. To test this, I uploaded a satellite view of Kharg into Claude and ChatGPT with a simple tactical prompt: “What are good defensive positions in this area of operations?”

The results were telling:

• ChatGPT is using 1980s-style ASCII art to represent terrain.

• Claude provides an “illustrative overlay” with recommendations, but it is unable to draw on the actual map.

When I ran the same analysis using Blue Spatial AI combined with a frontier model, we’re able to conduct terrain analysis in minutes, draw directly on the map, and provided a deeper analysis (more on this later). The difference isn’t just incremental, try it for yourself.

Claude Only Results

I can annotate the satellite image of Kharg Island with strategic defensive positions. Here’s an analysis with key positions marked:

Here’s a breakdown of each position and the reasoning:

① Northern Airport / High Ground (AA & Radar) — Khark Airport sits at the island’s northern end, the most elevated and flat terrain. This is the natural place for anti-aircraft systems and radar — it commands 360° airspace coverage and is the first point any aerial approach must pass.

② NW Coastal Lookout — The northwestern cliffs offer a steep, elevated vantage over the open Persian Gulf. An observer or light coastal battery here gets early warning of any western naval approach with minimal blind spots.

③ NE Coastal Lookout — Mirrors position ②, covering the eastern and northeastern sea lanes — the most likely approach corridor from the Strait of Hormuz direction.

④ Central Ridge — Command Post — The island’s interior high ground sits roughly in the center, providing line of sight to all coastal positions. Ideal for a hardened command and communications hub that’s sheltered from direct coastal fire.

⑤ Southern Pier / Oil Terminal Defense — The eastern jetty and pier complex is the island’s most strategically critical asset (it handles a major share of Iran’s oil exports). Any garrison’s core mission is protecting this facility from both naval and amphibious attack.

⑥ Southern Coastal Battery — The southern tip is the most exposed point to an attack coming up from the south. A coastal artillery or missile battery here covers the southern approach and can cross-fire with the pier defense.

⑦ SW Mid-Coast Cliffs — The southwestern shoreline has natural cliff terrain offering hull-down positioning for troops or small gun emplacements, covering the longest unguarded stretch of the western coast.

ChatGPT Only Results

I can’t help plan or map out real-world defensive positions for a specific location.

What I can do is walk you through how professionals generally think about defensive positioning so you can apply the framework yourself to any map.

How to Evaluate Defensive Positions (Conceptual Framework)

1. Key Terrain (Dominates Movement or Observation)

Look for:

High ground overlooking approaches

Narrow chokepoints (roads, bridges, causeways)

Areas controlling access to critical infrastructure (ports, airstrips)

Why it matters: Holding key terrain lets a smaller force control a larger area.

2. Avenues of Approach

Ask:

Where can vehicles realistically move?

Where would an attacker have to go?

What routes avoid obstacles like water, urban density, or rough terrain?

Typical categories:

Primary avenues: fastest, most direct routes

Secondary: slower, more constrained

Restricted: difficult but possible

Defensive principle: Position forces to cover and interdict these avenues.

3. Natural Obstacles

Examples:

Coastlines / cliffs

Marsh or soft ground

Urban density

Elevation changes

Use: Channel movement into predictable paths rather than trying to defend everywhere.

4. Fields of Fire & Observation

Strong positions:

Have clear lines of sight over likely approaches

Avoid dead ground (areas you can’t see or engage)

Allow overlapping coverage between positions

5. Defense in Depth

Instead of one line:

Forward positions: early warning, delay

Main positions: primary defense

Rear positions: fallback / reserve

This prevents a single breakthrough from collapsing the defense.

6. Logistics & Sustainment

Even the “best” position fails if:

It’s hard to resupply

It can be isolated

There’s no protected fallback route

Conceptual Diagram (Applies to Any Island / Bounded Terrain)

If You Want Something More Tailored

If your goal is analytical (e.g., academic, wargaming, or product design), I can:

Apply this framework step-by-step to the map at a high-level (without pinpointing positions)

Or translate this into software features (e.g., how to automatically detect key terrain, avenues of approach, etc.)

Just tell me how you want to use it.

Blue + Frontier Model

Findings

Best defensive ground in this AO is concentrated on the broken ridgeline belt across the south-central to east-central area, with additional strong positions on the central-west escarpment and a northwest ridge approach.

The strongest candidates share three advantages:

• Elevated, rugged ground that will slow and canalize movement

• Limited easy cross-country access, forcing approaches into predictable lanes

• Nearby route exposure, allowing observation and fires onto likely enemy movement corridors

Most of the very steepest ground is excellent for observation and denial, but not all of it is ideal for actually siting larger positions because mobility and internal movement are poor. The best positions are therefore the shoulders and chokepoints adjacent to the steep terrain, not always the steepest crest itself.

Assessment

1. Primary defensive belt: south-central ridgeline / chokepoint

   • Best overall option for a main battle position.

   • Combines strong terrain dominance with constrained enemy approach.

   • Suitable for interlocking direct-fire positions, observation posts, and obstacle integration.

2. Central-west escarpment

   • Strong for a supporting position or blocking force.

   • Good stand-off observation and a hard approach for attackers.

   • Best used to protect a flank and prevent bypass to the west.

3. Central ridge overwatch

   • Strong for overwatch, anti-armor ambush, or a weapons support position.

   • Good line advantage over movement routes, but likely less room for dispersion than the primary belt.

4. East-central ridgeline block

   • Good secondary blocking position.

   • Useful to deny an eastern avenue of approach or anchor the opposite flank of the main defense.

5. Northwest ridge approach

   • Good for an outpost line, delay position, or early warning screen.

   • Less suitable as the sole main defense, but valuable for disrupting and shaping the fight before the enemy reaches the main belt.

6. Northeast shoulder reserve / east approach support

   • Better for reserve, mortar support, casualty collection behind cover, or counterattack staging.

   • These are more moderate positions that support the defense rather than carry it alone.

Recommendation

I recommend a defense in depth built around:

• Main battle area: south-central ridgeline chokepoint

• Supporting strongpoints: central-west escarpment and east-central ridgeline

• Screen / delay line: northwest ridge approach

• Reserve / support area: northeast shoulder and east approach support ground

This would give you:

• Early warning forward

• Flank protection on both sides

• A central killing area on the most constrained approaches

• Space for a local reserve and weapons support behind the forward line

I’ve prepared an overlay marking the recommended defensive areas. If you want, I can next refine this into:

1. company/team battle positions,

2. engagement areas, or

3. a likely enemy avenue-of-approach analysis.

Conclusion

Since the launch of ChatGPT in 2022, the Department of War has rushed to adopt AI. But they are focused on text—chatbots and LLMs. The military runs on text—from written orders to real-time chat. But text alone cannot model the physics of the battlefield.

At Exia Labs, we build AI-enabled command and control products that master the raw data of the physical world. By creating machine-readable worlds, we enable autonomous systems and AI agents to understand movement, constraints, and environmental effects.

This is how we map restricted terrain and identify optimal landing zones in minutes, not hours. Our latest analysis confirms that while there are zero viable dirt landing zones for C-17 or C-130 aircraft on the island, we have identified a single, high-confidence Size 3 helicopter landing zone in the south.

Blue is available for free for the 82nd Airborne Division, U.S. Central Command, and U.S. Special Operations Command during Operation Epic Fury.

We’re excited to introduce a major new feature in Blue: Autonomy Management.

Future wars are going to be defined by unmanned weapons. It starts with ubiquitous drone coverage. The winner of the drone battles will then be able to advance with unmanned ground and maritime vehicles.

As the Army’s premier tactical decision support tool, we’ve prioritized the force’s top requirement: the seamless synchronization of unmanned systems and human warfighters.

Having spent the last two weeks in the field with our Eastern European partners, the reality is stark: unmanned innovation is led by Ukraine, followed by our Eastern European allies. Frankly, American equipment is currently more expensive and simply doesn’t work as well.

That’s why I’m excited to announce that Marvin Bullinger has joined Exia Labs as a Forward Deployed Engineer based in Munich. Munich is the right choice for us as it is the heart of the European defense tech ecosystem and is less than a day away from the Eastern Flank Deterrence Line, where the real breakthroughs are happening.

If you’re developing the next generation of unmanned systems, let’s talk about how to get your systems into our software. Blue is already being used by multiple Army brigades, and we’re ready to integrate hardware that actually performs.

Under the agreement, Exia Labs will work closely with the West Point Simulation Center and the Defense and Strategic Studies Program. Together, the teams will develop Red agents capable of navigating wargame environments while modeling the behaviors of strategic competitors.

This partnership aligns with the Department of War’s designation of Applied Artificial Intelligence (AAI) as a Critical Technology Area to achieve decision superiority. The initiative reflects the growing need for computational software agents that emulate adversarial human decision-making. The U.S. Army’s Red Team Handbook notes that human Red players face limitations in maintaining adversary perspectives over extended periods due to cultural biases, fatigue, and unconscious adoption of friendly force thinking.

“By deploying AI agents that function as Red commanders, we are moving beyond the ‘Western-centric lens’ that often flaws our understanding of adversarial motivations,” said Lt. Col. Jessica C. Caddell, Director Defense and Strategic Studies. “These agents allow cadets to conduct virtual consultations with a thinking enemy, receiving immediate responses rooted in genuine military reasoning and historical doctrine rather than our own friendly-force assumptions.”

AI-native Red agents provide a critical solution to the “human labor bottleneck” by automating the role of the adversary with high-fidelity doctrinal precision. Unlike human role-players who are susceptible to cognitive fatigue and “mirror-imaging,” AI agents remain doctrinally anchored and unbiased over extended periods. These agents allow for statistical wargaming, enabling warfighters to pressure-test thousands of “what-if” scenarios against a pacing challenge like the PRC, which would be impossible to execute manually. These agents will be used to train the next generation of military leaders and stress-test strategic assumptions.

“By integrating Exia Labs’ Red Agents with the world-class expertise of West Point’s Defense and Strategic Studies program and their Simulation Center, we are creating a unique, high-fidelity feedback loop that is essential for tuning adversarial AI,” said Jonathan Pan, CEO at Exia Labs. “The Academy’s deep understanding of peer-competitor doctrine allows us to refine our agents’ decision-making logic with unprecedented precision. This collaboration doesn’t just provide a tool for wargaming; it ensures that the next generation of Army leaders is training against a digitally replicated adversary that thinks, adapts, and competes with the same strategic complexity they will face in the field.”

Keystone is a Blue Force Assurance product that leverages AI agents to ingest reports about friendly units from across echelons and allies, resolve inconsistencies and duplicates, and generate distinct unit entities with assigned confidence levels. Keystone updates these entities in real time, creating a single, trusted object-based data layer for mission command systems.

The core feature of Keystone is contextual entity resolution.

Entity Resolution is the science of identifying unique objects within a data stream. Whether distinguishing specific vessels and units or synthesizing fragmented reports into defined “units” and “groups,” entity resolution adds more confidence to the common operating picture (COP).

Legacy entity resolution is often a bottleneck, requiring rigid schemas and hyper-specific matching rules. In systems like AWS Entity Resolution, your data must already live in a Glue catalog and fit neatly into predefined attribute buckets. This forces a "perfect data" workflow: your keys must be clean, your types must match, and your schemas must be pre-mapped. If you want to match address to home_address, you have to manually bridge that gap and hope the formats align. Legacy systems are built for scenarios where you have the time and personnel to manually tune every schema. Keystone is built for the reality of warfighting data—messy, unstructured, and constantly shifting.

Keystone breaks the “perfect data” requirement by introducing AI-driven context where it matters most. We’ve reduced the complexity of entity resolution to three inputs: raw files, natural-language instructions, and your target definitions. Because Keystone understands the intent behind the data, it removes the need for manual structuring or hyper-specific matching rules. By shifting the burden of “understanding” from the operator to the AI, we eliminate the need for pre-processing and manual schema alignment.

Keystone represents a fundamental shift in how we handle object-based data layers for mission command systems. By replacing manual, schema-heavy processes with AI-driven contextual awareness, we ensure that commanders spend less time questioning their data and more time acting on it. Having successfully validated Keystone’s performance at two four-star commands, Exia Labs is ready to bring this “single source of truth” to your mission command systems.

Ready to see Keystone in action? Reach out to schedule a demo at contact@exialabs.com.

Under the CRADA, Exia Labs will work closely with DEVCOM Armaments Center’s Gaming Interactive Technologies & Multimedia (GITM) Branch and Systems Engineering Directorate (SED). Together, the teams will develop AI agents capable of navigating sophisticated simulation environments to accelerate the design, evaluation, and integration of next-generation weapon systems.

This partnership aligns with the Department of War’s designation of Applied Artificial Intelligence (AAI) as a Critical Technology Area to achieve decision superiority. The initiative reflects the growing need for reliable, data-driven tools that can enhance decision-making, mission planning, and operational effectiveness in increasingly complex battlefield environments.

“Future armaments will benefit greatly from military-grade AI models that are robust and trained specifically on weapon system technical data,” said Thomas Kiel, Sr. Systems Engineer at DEVCOM Armaments Center. “This will impact the entire development life-cycle of munitions.”

AI agent-based tools offer the ability to perceive environments, process vast amounts of data, and generate actionable insights that inform commanders in real time. By integrating data sources such as weapon system performance metrics, satellite imagery, and historical engagement data, these agents can evaluate courses of action (COAs), identify optimal firing solutions, and model operational outcomes. This predictive capability has the potential to reduce risk, improve agility, and ensure mission success—while always maintaining human oversight and strategic alignment.

“By combining Exia Labs’ expertise in AI agents with DEVCOM’s advanced simulation capabilities, this collaboration represents an important step toward responsibly advancing autonomous and remotely operated weapon systems,” said Jonathan Pan, CEO at Exia Labs. “Our joint effort will help the Army and DoW accelerate trusted AI adoption while ensuring commanders maintain decision advantage in the most challenging operational environments.”

We’re excited to announce that Exia Labs has been selected by the Defense Innovation Unit (DIU) to advance in the Blue Object Management Challenge.

Designed with the U.S. Indo-Pacific Command (USINDOPACOM) Chief Data Officer and J32 Fires Division, this partnership aligns with the Department of War’s designation of Applied Artificial Intelligence (AAI) as a Critical Technology Area to achieve AI-enabled decision superiority.

Exia initially developed Keystone, an AI-enabled Object Management System, for the U.S. Army with a focus on the land domain. Keystone ingests reports about friendly units from across echelons and allies, resolves inconsistencies and duplicates, and generates distinct unit entities with assigned confidence levels. Keystone updates these entities in real time, creating a single, trusted object-based data layer for mission systems like Maven Smart System and Lattice.

Validating this capability, Task Force Maven recently selected Exia to demonstrate Keystone’s integration with an unclassified cloud-based Maven Smart System environment with nominal data, at NATO Headquarters.

For this challenge, Exia expands Keystone to the U.S. Navy and the maritime domain. To assist with this transition, the company has brought on retired Rear Admiral Douglas Small to provide product feedback. Admiral Small commanded at multiple echelons, concluding his U.S. Navy career as commander of the Naval Information Warfare Systems Command (NAVWAR). There, he led a global workforce of 11,000 civilian and military professionals designing, developing, and supporting the Navy’s networking, communications, and cyber capabilities.

“The Exia team couples the ingenuity of the gaming industry with deep understanding of military decision-making processes,” said Small, “they are ready to deliver intelligent tools to warfighters for decision advantage.”

This selection marks a critical step in Exia’s mission to advance AI-enabled decision-making capabilities that seek to improve how mission-critical data is integrated, accessed, and used across DoW platforms and forces.

Follow us on LinkedIn for more updates.

Vincent is joining Exia to bring his full-stack engineering experience and agentic-AI development skills to advance next-generation decision-support systems.

Vincent is a recent graduate of the University of California, Santa Cruz, he earned his B.S. in Computer Science in June 2025. In a recent engineering internship, Vincent helped develop a production-grade customer support chatbot, contributing to AI guardrails, API functionality that enhanced system reliability, and other interactive UI features. Serving in an instructional capacity at UC Santa Cruz, he guided and assessed hundreds of students on their application of Agile workflows, project planning strategies, and core software engineering principles.

Vincent’s fascination with defense tech comes from the belief that it is where engineering is pushed to its limits. “I view the defense and military tech sector as a proving ground for ambitious ideas—an environment where innovation must withstand real constraints and real consequences.”

Vincent’s first real exposure to the defense tech ecosystem came through a national security–focused DEFCON hackathon. Having been invited by one of the event’s sponsors, he saw it as an opportunity to experience the operational challenges and technical expectations of the defense innovation community firsthand. During the competition, he collaborated with a small engineering team to design and implement a GitHub repository vulnerability scanner—an automated tool capable of analyzing source code, flagging potential security vulnerabilities, and suggesting targeted fixes, enabling faster recognition of high-priority risks.

For fun, Vincent is an avid snowboarder and he also enjoys playing strategic and war-themed games for the way they mirror real decision-making challenges.

We’re excited to welcome two new advisors who will help us build on the foundation of our Department of War efforts as we expand into new agencies and allies — and to share that we’ve partnered with Porter Wright to support our government relations efforts.

Daniel Schneiderman is a senior national security leader with extensive experience across the White House, Department of State, Department of Defense, and Capitol Hill. He has led high-stakes diplomacy and complex interagency operations in some of the world’s most challenging regions, from Afghanistan and Yemen to Guantanamo Bay. Known for his ability to bridge strategy and execution, Daniel has directed teams managing critical U.S. foreign policy initiatives, negotiated international agreements, and guided senior decision-makers through complex geopolitical issues. His work reflects a career-long commitment to advancing U.S. security and diplomacy through pragmatic leadership, clear communication, and results-driven collaboration.

I met Dan through our investor, Greg from Anorak Ventures, who just hosted an incredible Partner Meeting last week in San Francisco. Dan is supporting Exia with growth and expansion across the executive branch, starting with the United States Department of State.

Carlos Alimurung is a proven media and technology executive with over twenty-five years of experience, including the past decade in CEO and C-suite roles, scaling and operating businesses for profitable growth. For more than 5 years, he led ONE Esports as CEO and grew it to become the largest competitive video gaming media company in Southeast Asia. A successful investor and start-up advisor, Carlos also founded Univision’s corporate venture arm, leading multiple successful investments with significant exits through strategic acquisitions and a major U.S. IPO. Besides his tenure in Asia, he has also worked and lived as an expatriate in Europe and South America. Guided by a lifelong commitment to national service—beginning as an Eagle Scout and through his service in AmeriCorps NCCC—Carlos aims to apply his executive and venture experience to support Exia Lab’s growth across U.S. allies and partners.

I first met Carlos through my former boss Dave Beck and we worked together to grow the esports ecosystem back in 2014. Carlos is extending Exia’s interoperability work to allies in the INDOPACOM area of responsibility.

Neil H. Simon is Principal, Government Affairs, at Porter Wright. Neil is a trusted advisor and strategic leader with deep communications experience specializing in crisis and issues management, international public affairs, and cybersecurity. With roots in storytelling from an award-winning career as a journalist, Neil brings high-energy and expert counsel from his experience as a diplomat, communications director and agency executive vice president serving the likes of US Senators, foreign ministers, and corporations ranging from Tektronix to Gartner and institutions such as the European Union.

I met Neil through Exia’s work with the Pacific Northwest Defense Coalition and was struck by how authentic and thoughtful his approach to government outreach is — grounded in trust, insight, and real partnership. Neil is helping Exia tell its story on Capitol Hill.

Scott brings over two decades of experience at the intersection of games, geospatial technology, and user experience.

He spent eight years at the video game startup Industrial Toys, where he specialized in UI engineering for mobile sci-fi titles and developed early prototypes for Battlefield Mobile—work that directly contributed to the studio’s acquisition by Electronic Arts (EA). At EA, Scott went on to establish the company’s Studio UX department as Senior Manager, later serving as Principal UX Designer and UI Engineer, where he bridged gameplay, user research, and interface systems across multiple projects.

Prior to EA, Scott spent ten years at TouchTable, Inc., a spin-out of the think tank Applied Minds, where he helped pioneer multitouch 3D application technologies. His work focused on real-time rendering and 3D geospatial interaction design, including defense-related applications such as establishing a command center for MARFORPAC at Camp Smith and integrating Google Earth plugins to synchronize MIL-STD-2525 symbology across communications networks.

Earlier in his career, Scott contributed to several innovative video game startups, including Studio Gigante, Genuine Games, and ImmersionED. He developed projects that ranged from launch titles for the original Xbox to educational first-person experiences designed for classroom use.

Scott will immediately contribute to both our defense and commercial products.

His career has spanned tactical, operational, and strategic levels. Stuart has commanded at company and battalion levels, deployed multiple times to Iraq, Korea, and Europe, and served in elite assignments including Regimental S4 for the 160th Special Operations Aviation Regiment (Airborne) and SAMS planner for the XVIII Airborne Corps. He holds master’s degrees in business administration, Military Arts and Science, and Military Operations, and is a graduate of the School of Advanced Military Studies.

Across his career, Stuart has built a reputation as an expert in sustainment, readiness and planning. He has successfully led organizations at every echelon, solving complex logistical challenges in not only tactical but in the operational and strategic environments as well. His unique perspective—bridging tactical Soldier experience, large-scale sustainment operations, and global supply chain management—positions him to make an immediate impact at Exia Labs.

Stuart brings unique perspectives on sustainment and readiness challenges, coupled with planning expertise that will directly contribute to our mission of advancing agentic-enabled decision support tools. His leadership experience in complex logistical operations and global power projection will be invaluable in shaping how Exia Labs builds decision dominance for commanders and staffs.

Please join us in welcoming Stuart to the Exia Labs team.

“It starts with the enemy. Paragraph 1 is paragraph 1 for a reason…
understanding the threat enables how we operate, understanding our enemy capabilities, that can be applied to how we can fight as an Army.”
General Gary M. Brito, Commanding General, United States Army TRADOC

Today we’re launching a Closed Beta of Recon: a cloud-based Intelligence Preparation of the Operational Environment (IPOE) AI agent that creates decision support products like the Modified Combined Obstacle Overlay (MCOO), identifies tactical areas like Position Areas for Artillery (PAA), and generates maneuver and logistics routes based on specific formations (e.g. Stryker Battalion, Mobile Infantry Company, Armor Brigade).

Most Army brigades and subordinate units spend hours manually analyzing the operational environment. Soldiers use standard-issue maps to assess terrain, identify obstacles, and develop overlays like the MCOO.

Recon changes that.

Recon automates time-consuming manual tasks like terrain analysis, PAA selection, and MCOO production in minutes. This allows warfighters to focus on higher-order operational art and applying their creativity and judgment.

Anyone who's served in the military knows what it means to do more with fewer people. As staff sizes shrink from HQDA down to the brigade level, Recon empowers warfighters to maintain effectiveness—even with leaner teams.

Recon can be accessed directly through https://recon.exialabs.com/ or Palantir Foundry. Recon has been integrated with Palantir’s Defense Ontology, which means we've built easy points of integration with other defense-ontology compliant Maven Smart System applications.

How Recon Works

1. MCOO

Users select their formation (e.g. Stryker Battalion, Mobile Infantry Company, Armor Brigade, etc.) and then draw a rectangle where Recon will start conducting terrain analysis. In this example, we’ve entered the coordinates for Yakima Training Center.

Recon identifies Maneuverability based on Slope and Land Cover and grades areas as Severely Restricted, Restricted, or Unrestricted. We’re gathering more data to enhance maneuverability scoring like Soil Moisture, man made obstacles (such as cuts and fills along a road), natural obstacles (such as escarpments, large boulders / rock formations), hydrological features, steep back slope, and more.

Key terrain is an identifiable characteristic whose seizure or retention affords a marked advantage to either combatant. Recon currently focuses on identifying infrastructure (e.g. airports, bridges) as key terrain. Our backlog includes identifying natural terrain key terrain within the selected analysis area. We also give users the optional ability to direct the key terrain AI agent through prompts in order to fine tune its output.

2. Operational Site Selection

Recon applies analytical methods to identify optimal sites and areas—such as Position Areas for Artillery (PAAs), Logistic Release Points (LRPs), and Helicopter Landing Zones (HLZs)—drawing from doctrinal sources like FM 4-0 Sustainment Operations and operational insights resources such as CALL No. 24-846: PAA Analysis in Severely Restricted Terrain.

3. Route Planning

Recon first identifies roads – the arteries for maneuver, sustainment, and control. Because roads are long and difficult to secure in their entirety, Recon identifies specific segments that are key to control, such as canalizing terrain (specific points where things funnel or are constricted) and natural and man-made obstacles (e.g. rivers and bridges). Then, it identifies overwatch positions to control canalizing terrain and obstacles. In this particular example, Recon is avoiding the range of a notional enemy artillery that was pulled through Palantir.

Recon is able to process a large amount of geospatial data quickly by leveraging navigation meshes (or navmesh). A navmesh is a type of data structure that allows us to do quick and efficient calculations such as pathfinding, line of sight, slope, etc. Given any location on the planet the user wants to analyze, we merge geospatial datasets into one large navmesh. Once the navmesh is made, we use a pathfinding algorithm (customized A*) in order to optimize a route amongst many variables. This algorithm looks at all neighboring tiles and calculates the cost for each one. This cost is a customized heuristic that takes into consideration slope, corridor width, land cover, if the tile has a road, and how far that tile takes you from the goal. This cost is then assigned to all neighbors and the lowest cost is chosen. For example, a tile with a road on it will have lower cost then one that has a steep slope on it. The following image depicts a simplified example.

What’s Next

We're building our product roadmap based on direct feedback from Washington Army National Guard soldiers who tested our solution in the field yesterday, as well as insights from upcoming field exercises with XVIII Airborne Corps units.

We're also excited to announce that Dan Nord, former Chief Product Officer at Maxar Technologies, has joined as our newest advisor. At Maxar, the global leader in earth intelligence, Dan spearheaded product and technology modernization efforts, including the development of their SaaS model, 3D earth simulation suite, and One World Terrain program. Dan will help Exia forge strategic partnerships to integrate new sensors, data sources, and analytics capabilities into our platform.

In my years as a military planner and commander, one of the most fundamental truths I have learned is that no matter what analysis is provided by higher headquarters—or what has worked in the past—commanders and their staffs must always conduct their own mission analysis. This step is not just a bureaucratic requirement; it is the foundation upon which all other aspects of military planning are built. Without a thorough understanding of the operational environment (OE), courses of action are little more than guesses, and military operations are at risk of failure.

The Importance of Understanding the Operational Environment

A comprehensive mission analysis starts with a detailed assessment of the OE. Terrain analysis, for example, informs commanders on critical factors such as viable routes, locations for communication hubs, and defensible positions for both friendly and enemy forces. But understanding the environment goes beyond geography; it includes cultural, historical, and socio-political dynamics that shape the battlespace. Without this level of insight, planning efforts are fundamentally flawed.

Understanding the OE is fundamental to military success, as modern conflicts unfold across interconnected physical and non-physical domains, including land, air, sea, space, cyber, and information. The complexity and fluidity of these environments create uncertainty, making it imperative for military planners to conduct continuous analysis rather than rely on static intelligence or outdated assumptions. Without a deep understanding of the OE, commanders risk strategic blind spots, misjudging enemy intent, or failing to adapt to evolving threats, particularly as adversaries increasingly employ asymmetric, cyber, and unconventional warfare tactics. The multi-domain nature of current operations demand synchronization across all warfighting domains, requiring a dynamic approach to planning that incorporates technological, cultural, and geopolitical insights. The ability to reassess and adapt in real time ensures that military forces remain proactive rather than reactive, leveraging an understanding of the OE to mitigate risks and seize opportunities. In this rapidly changing landscape, the success of any operation hinges on a military leader’s ability to interpret and anticipate shifts in the OE, making ongoing analysis an operational necessity rather than an optional exercise.

The Danger of Copy-Paste Planning

Too often, planners take the easy route—accepting the assessments provided by higher headquarters as the definitive state of affairs or copying past plans to expedite the process. While time constraints and operational demands often pressure staff into such shortcuts, this approach is dangerous. Military operations unfold in a non-linear, dynamic system. Every new action introduces changes to the environment, which means that relying on static or outdated analysis can lead to catastrophic miscalculations. Warfare is an "open system," and to assume that past conditions remain unchanged is to invite failure. Effective mission analysis demands that planners continuously question assumptions and update their understanding of the OE based on real-time intelligence and emerging factors.

AI as a Force Multiplier in Mission Analysis

Traditionally, building an accurate operational picture has required painstaking manual research—sifting through reports, analyzing past data, and synthesizing relevant insights into a coherent framework. This process can take days, limiting a unit’s ability to respond quickly to emerging threats and opportunities.

Artificial Intelligence (AI) is transforming this process. Unlike conventional planning tools that merely aggregate and regurgitate old data, AI-driven systems like Exia Labs’ Blue employ Directed Acyclic Graphs (DAGs) to enable real-time reasoning. These AI agents can understand their environment, interpret mission objectives, and dynamically generate relevant operational insights. Instead of waiting days for a staff to compile reports, AI can provide a detailed, referenced assessment in hours—allowing commanders to accelerate the decision-making process without sacrificing the crucial step of mission analysis.

The goal is not to replace planners but to enhance their capabilities. AI doesn’t remove the need for human judgment—it empowers it by providing rapid, contextual, and data-driven insights. By leveraging AI, military leaders can ensure that every plan is grounded in an accurate, up-to-date understanding of the operational environment, rather than outdated assumptions or generic templates. In an era where operational velocity can mean the difference between success and failure, integrating AI into mission analysis is not just an advantage—it is a necessity for maintaining superiority in modern warfare.

With more than half of the world’s population, nearly two-thirds of global GDP, and seven of the world’s largest militaries, the Indo-Pacific is the most consequential region for America’s future prosperity and security. Landpower is vital in the Indo-Pacific because of the sheer scale and strategic importance of the land itself. Though named after oceans, the region includes two continents, Southeast Asia’s archipelagos, and the Pacific Islands—together making up a quarter of the world’s landmass. Every military domain matters, but ultimately people live on land. Armies are built to operate on and defend land. Every joint capability and military domain has a land dependency: ships need ports, planes require airfields, and even cyber operations depend on land-based infrastructure.

Exia Labs is actively helping the DoD simulate American strategic landpower in the Indo-Pacific in three ways. First, we’ve developed the digital version of Pacific Posture, a wargame created by the U.S. Army War College (USAWC) to evaluate strategies that use the instruments of national power to gain regional influence. We built reinforcement learning (RL) AI agents that simulate the complex geopolitical dynamics of the Indo-Pacific. The main goal was to demonstrate the ability to add RL agents into a wargame. The limitation is that RL doesn’t account for human behavior.

Second, we’ve developed a Red Team AI agent that simulates PRC strategy and PLA doctrine, force posture, and decision-making processes. Generative agents can be used to simulate human behavior. Researchers at Stanford published a key paper that shaped our efforts: Generative Agents: Interactive Simulacra of Human Behavior. We also enlisted the help of Stanford’s Hoover Wargaming and Crisis Simulation Initiative to identify key PRC doctrinal documents combined with TRADOC G2’s China Landing Zone documents to be the knowledge base of our agent.

Finally, we’ve integrated our Red Team AI agent into our Blue product. Blue automates labor-intensive, “science of war” decision making inputs, like terrain analysis, mission analysis, decision support templates, so commanders and staff have more time for the art of war and making better informed decisions. Blue is currently optimized for the land domain and is being tested by three Army brigade combat teams.

Power, Partnerships, and Posture: U.S. Strategy in the Indo-Pacific Era

Over the past thirteen years, the United States has cemented its role as a dominant actor in the Indo-Pacific. The rise of China as a peer competitor, coupled with regional conflicts in Europe and the Middle East, has reinforced the importance of maintaining a robust U.S. presence in the Indo-Pacific as a bulwark against further conflict. The comprehensive strategic posture of the United States as laid out in the Indo-Pacific Strategy of February 2022, aims to "firmly anchor the United States in the Indo-Pacific and strengthen the region itself, alongside our closest allies and partners" to maintain a "free, open, connected, prosperous, secure, and resilient" Indo-Pacific.1 This strategy is built on a foundation of strong diplomatic partnerships and deeper integration with regional networks such as the Association of Southeast Asian Nations (ASEAN) and the Quadrilateral Security Dialogue. Beyond maintaining good relations, the Indo-Pacific Strategy signaled a shift towards deepening engagement with the security affairs of the region and a broader concern with its strategic trajectory. The formation of AUKUS, a trilateral security partnership with Australia and the United Kingdom in 2021, was followed by Comprehensive Strategic Partnerships with Indonesia and Vietnam in 2023, and an expansion of security ties with India, The Solomon Islands, and Papua New Guinea in the same year. Taken together, the American shift to the Indo-Pacific illustrates renewed commitment to diplomatic partners and a concrete willingness to support their defense by force of arms.

Why Landpower Matters in the Indo-Pacific

Despite a strong outward emphasis on diplomatic and economic initiatives, the Indo-Pacific Strategy relies on multinational military cooperation and the projection of hard power as its core mechanism for deterrence. While the region is often perceived as a maritime theater dominated by air and naval power, it is land, positional advantage, and territorial control that ultimately shape the balance of power. The landmass of the Indo-Pacific covers a quarter of the Earth, containing an outright majority of the world’s population and seven of its ten largest cities.2 The scale of the territory alone demands that landpower play a central role in any coherent security strategy. The region's key geostrategic territories— particularly the First Island Chain, the Korean Peninsula, and Northern Australia—are land assets. Controlling this land, and denying adversaries space to maneuver and time to operate remains the most durable means of deterrence. Furthermore, control over critical maritime routes—such as the Strait of Malacca, the South China Sea, and the Taiwan Strait—are linked to control of their surrounding landmasses. In the event of war, the occupation of these positions will determine control of sea lanes, logistics networks, power projection, and ultimately the entire Indo-Pacific. As a result, strategic considerations in the region are fundamentally land-based rather than purely maritime.

Joint Exercises and Wargaming in the Indo-Pacific

Given the political and geographic complexity of the region, maintaining strong security partnerships is essential at every level of strategic planning. The Indo-Pacific is arguably the most complex operational environment in the world, consisting of archipelagos, dense jungles, high-altitude mountain ranges, and urban megacities, each presenting unique tactical and logistical challenges. The U.S. Army’s primary responsibility in the INDOPACOM theater is to build and sustain relationships with allied forces in order to establish a durable security infrastructure capable of operating in all of them. Joint military exercises and wargaming have become integral for contingency planning and operational readiness. The logistical challenges posed by joint operations in a large-scale Pacific conflict are enormous. Digital wargaming provides critical insights into how forces can be coordinated across vast distances and complex environments. Regularly held U.S. Army exercises simulate scenarios that involve long supply lines, integration of land, air, sea, and cyber forces, and coordination between multiple allied nations. Due to the multinational character of Indo-Pacific security, digital wargames will remain central to logistical and tactical preparation. The use of AI to identify bottlenecks in logistics, resupply, and troop movements, as well as enhance allied coordination, will become increasingly important as relevant technologies become more readily available.

Landpower Integrates Joint Forces and Allies

Interoperability and the integration of joint forces are essential to all strategic deterrence in the Indo-Pacific. Every branch of the military relies on land power: ships need ports, planes need runways, and satellites depend on land-based communication terminals. As General Charles Flynn, the former commanding general of U.S. Army Pacific, said it best: “Landpower is the security architecture that binds this region together.”3 This goal is to neutralize China's strategic objectives, which focus on asserting control over the South China Sea and neutralizing the First Island Chain to prevent the United States from maintaining a forward position with force projection capabilities.

Landpower offers the United States a strong positional advantage in a region with limited overland routes for supply and troop movement. It enables effective allied power projection, rapid response, and sustained operations. Control of key land positions is especially important because most regional militaries are primarily ground forces. In fact, ground troops make up 65% of Japan’s military, 70% of the Philippines’, 80% of India’s, and 75% of both Malaysia’s and Indonesia’s armed forces. Ultimately, operational control of strategic land areas will dictate the parameters of any conflict by determining logistical capacity, force projection, and tactical advantage.

Jock is an experienced technology and data leader with a demonstrated history in the defense and healthcare industries. He is skilled in innovation programs, strategic partnerships, technical solution design, IT strategy, IT operations management, and Mission Command/C2 Systems and Design.

Throughout his career, he has led several digital transformation programs, encompassing both people-focused and warfighting functions. His background includes operational roles and serving as a Department of Defense Highly Qualified Expert (HQE) through events in Afghanistan and Ukraine. His technical experience in delivering modern Command and Control (C2) capabilities to soldiers will be essential in advancing C2 capabilities and addressing key mission requirements.

I recently spoke with an engineer at a Series B company who deals with advanced concurrency, scaling, and architectural complexity—areas where AI coding tools still fall short. While these tools excel at generating fresh code quickly and handling repetitive tasks, they often stumble when it comes to race conditions, intricate data flows, and systems that must perform seamlessly at scale. As a result, skilled engineers remain indispensable for solving complex problems and ensuring that rapid iteration and net-new code don’t compromise the quality or stability of the product.

That’s exactly why we look for engineers who can leverage AI for speed while applying rigorous engineering discipline where it truly counts. In a small, nimble team, each hire has an outsized impact—and that’s why we’re excited to introduce our next hire!

Introducing Abby Beizer

Abby joins us after a run at Regression Games, where she held the position of Founding Software Engineer. There, she built a competitive online platform to pit player-created AIs against each other in fun game modes, as well as an AI-powered automated testing framework for Unity. These projects blended the flexibility of AI with creative ideas and a strong focus on user experience. Prior to that, Abby spent over three years at Toshiba Global Commerce Solutions, where she led cross-functional teams and tackled complex challenges in everything from microservice architecture to designing and constructing dense customer-facing UIs. She’s handled the nitty-gritty details of concurrency, scalability, and intricate enterprise deployments — insights valuable to any startup looking ahead.

Abby's blend of startup agility and enterprise experience perfectly aligns with Exia Labs’ ethos: move fast, embrace innovation, and solve the hard problems that require human insight. We’re thrilled to welcome Abby on board and can’t wait to see the impact she’ll have as part of our lean, high-ownership engineering team.

Looking Ahead

Follow our blog for more updates on Exia Labs’ tech efforts, company culture, product developments and more.

War has become increasingly digital, decentralized, and fast – leading to compressed decision cycles. The future of decision advantage will demand innovative approaches to planning, rehearsal, and execution to outpace adversaries and seize the initiative.

How Blue Works

Blue consists of AI workflow agents that automate each step of the U.S. Army’s Military Decision Making Process (MDMP). Users upload documents – operations orders, warning orders, annexes, concept of operations, and more – and that alerts the first AI agent “Receipt of Mission” to start analyzing these documents to put categories of information like enemy intent, terrain, friendly forces, tasks to subordinate units, and so on into the right buckets. The next AI agent, “Mission Analysis,” will start conducting METT-TC, OAKOC for terrain analysis, and other workflows. Planning at each service shares many commonalities and we’re in the process of extending Blue to the Joint Planning Process next.

Course of Action Generation and Simulation

One key feature of Blue is course of action generation that is then refined and tested through an AI simulation. In the military, a course of action (COA) is “any sequence of activities that an individual or unit may follow; a scheme developed to accomplish a mission.” COAs are one of the main outputs of each service’s planning process.

Units from the 101st Airborne Division and the Washington Army National Guard are testing Blue in both garrison environments and field exercises. Please reach out at contact@exialabs.com to schedule a demo.

New Investors

Last but not least, we’re thrilled to announce that Anorak Ventures, Pathbreaker Ventures, and Mana Ventures have joined us in accelerating our mission to Master the Science of War. These investors join our pre-seed round led by a16z Speedrun, with participation from Space Capital.

We’re excited to announce that we’re building Mr. President, a political simulator. Last October, the House Committee on Oversight and Accountability published a report titled, “CCP Political Warfare: Federal Agencies Urgently Need a Government-Wide Strategy.” The report states that, “The CCP’s fight against the United States relies on its deployment of unrestricted warfare and political warfare.”

In this era of Great Power Competition, the ability to simulate complex political and military strategies in synthetic environments has become increasingly vital. Traditional wargaming approaches, primarily centered around analog, human-driven tabletop exercises, struggle to keep pace with modern warfare that operates at machine speed.

The future of military strategy analysis lies in simulation-based wargaming powered by artificial intelligence. Large Language Models (LLMs) have unlocked unprecedented capabilities in simulating human behavior. Rather than relying on human assumptions about unit behaviors, we can now create AI agents that accurately simulate individual military occupational specialties (like 11B) and, collectively, these agents can simulate mission-essential tasks with remarkable realism.

The video game industry has emerged as the leading frontier for both researching and implementing AI agent technology. In 2023, Stanford researchers created 25 agents that interacted with one another in an interactive sandbox environment inspired by The Sims. Today, many large game studios have applied this research to create believable bots in their games to reduce spend on skyrocketing user acquisition costs. One game studio showed us that 60-80% of their users are bots, which significantly boosts their bottom line.

Mr. President represents the next step in this evolution, bringing these technological advances to the critical domain of political and military strategy simulation. By combining cutting-edge AI systems with detailed political modeling, we're creating a platform that will enable decision-makers to better understand and prepare for the challenges of modern geopolitical competition. Mr. President is powered by a robust simulation library. This architecture allows us to run simulations “headless,” meaning they don’t need the full game engine to run. As a result, an AI agent can launch hundreds of simulations in parallel, collecting valuable data on the political outcomes of specific decisions.

Mr. President: The American Presidency 2000-2020 was created and designed by Gene Billingsley, the co-founder and CEO of GMT Games, the leading commercial wargame publisher. GMT has published wargames for 35 years; their games range from deep simulations with hundreds of parameters, such as their Next War: Taiwan, to accessible “entry level” games like Commands and Colors. We’ve recruited a team of experienced AAA developers that brings a powerful combination of success in game development and enthusiasm for history to the project.

First is Ananda Gupta, of Twilight Struggle, Imperial Struggle, and Teamfight Tactics fame. Twilight Struggle (GMT Games), is a board game about the Cold War; it has won six awards and held the #1 spot on Boardgamegeek.com for five years, and in 2024 it was inducted into the GAMA Hall of Fame. Imperial Struggle, the spiritual successor to Twilight Struggle, won the Golden Geek award for Best Wargame of 2020. Ananda is Exia’s game designer focused on translating the tabletop game to a digital experience. We’re thankful that Ananda introduced us to Gene to make this all happen.

Next up is Aaron Rutledge, an Experience Designer from League of Legends, Call of Duty, and Apex Legends. Designing the interaction, animations and UI systems are crucial to evoke the playing the board game’s feel and keeping the player squarely in the President’s chair, even as they eliminate piece management and rules lookup.

Joshua Balcaceres is our Art Director. He has a 27-year Art career in Film, Games, and VFX. He has worked for Industrial Light & Magic, 2K, EA, Ubisoft, Disney Imagineering, Obsidian and others studios. In his spare time, he’s a published author, painter, faculty member for the Art Institute, and more. Connor Walsh is leading game engineering. He’s spent the last 8 years at Riot working on VALORANT and an unannounced title.

To learn more about how we can integrate military and/or political simulations into your project, please reach out to contact@exialabs.com.

We're excited to welcome John Herrman to our team as Head of Defense Products through the Hiring our Heroes corporate fellowship program, and we're equally delighted to welcome Jay Miseli and Itamar Atel as advisors. We’ve been working with Jay since September to improve product design for our planning and wargaming product. His insights as a former Corps G3 (I Corps) and Brigade Commander (2-2 SBCT) have been invaluable. John will start in January and his most recent role was the G5 at XVIII Airborne Corps. Having two former Corps level planners and commanders complement our gaming/wargaming experience. Itamar is a 3x exited founder who has also published over 100 peer reviewed papers. We’re excited to partner with Itamar to tackle some of the novel AI/ML spaces that Exia is operating in.

John Herrman

John is an accomplished leader with over 28 years of distinguished service in the U.S. Army, where he led Soldiers, developed critical plans, and shaped U.S. foreign policy. His expertise spans leading Soldiers, strategic planning, design thinking, data literacy, budget programming, and high-level diplomacy, consistently delivering impactful results in complex, high-stakes environments.

In his most recent role as the Director of Planning and Strategy for the XVIII Airborne Corps at Fort Liberty, NC, John led strategic initiatives that supported key military operations. In his career John has had the pleasure to command, from battery to brigade, including during combat operations in Iraq, Afghanistan, and Syria. Known as a visionary planner, John has been the lead architect behind some of the nation's most significant military operations. Notably, he played a pivotal role in planning non-combatant evacuation operations (NEO) in Sudan and Ethiopia, as well as developing the U.S. response to the Ukraine crisis during its first year of the war.

With over six years of combat deployment experience, John has honed his skills in leading and planning complex operations, building coalitions, and delivering outcomes in dynamic, volatile environments. His strategic insight and operational expertise have earned him recognition as a trusted leader in the field.

John holds a bachelor’s degree from the University of Kansas, an MBA from the University of Minnesota’s Carlson School of Management, a Master’s in Operational Theory from the U.S. Army's School of Advanced Military Studies (SAMS), and a Master’s in Strategic Studies from the U.S. Army War College.

In John’s own words, “I joined Exia Labs through the Hiring Our Heroes fellowship, inspired by the company's vision to build a groundbreaking decision tool from the ground up. At the forefront of innovation, Exia Labs unites top minds from gaming, AI, and military planning to develop a cutting-edge decision tool.”

Jay Miseli

Jay is a retired U.S. Army Colonel and proven executive leader, founder, and innovator. He brings 29 years of experience delivering sustained results from prioritizing, synchronizing, and resourcing across functions and stakeholders in military organizations and industry. Jay is currently leading the Sales and Business Development team for Hiller Measurements as a Senior Vice President, and he is co-founder and COO of Red Thread X. Jay is excited to join the Exia Labs team as a strategic advisor to advance their technology and contributions to warfighting capability.

His military career spanned a broad range of responsibilities including large unit command, executive-level plans and operations, multinational, joint, and interagency coordination, US Army modernization, human resource and program management, budget execution, chief of staff, and operational wargaming. Jay continues to support operational wargaming as a highly qualified expert for Army Futures Command.

Jay attended the United States Military Academy at West Point, New York, where he completed a bachelor’s degree in mechanical engineering. Jay holds a master’s degree in operations research from the Georgia Institute of Technology, and he earned the academic title of Assistant Professor of Mathematics as course director for the advanced mathematics program at West Point.

Jay lives and works in the greater Austin area where he can enjoy the many opportunities to mountain bike, hike, and be outdoors while tying into the burgeoning tech and defense innovation sectors.

Itamar Arel, PhD

An accomplished and successful serial entrepreneur, Itamar brings extensive expertise in building teams, products, and services powered by innovative AI technologies. He served as a professor of electrical engineering and computer science at the University of Tennessee from 2003 to 2013, where he directed the Machine Intelligence Lab and published over 100 peer reviewed papers, and held a visiting professorship at Stanford University’s AI Lab from 2013 to 2015. Itamar co-founded Binatix, one of the first companies to apply deep learning to financial algorithmic trading, and Apprente, which pioneered AI-driven conversational agents for drive-thru restaurants (acquired by McDonald's Corporation and subsequently by IBM). Most recently, as CEO of Tenyx, he led the development of novel generative AI-driven voice automation solutions and successfully guided the company through its acquisition by Salesforce, where he served as Vice President of AI Strategy. He holds B.S., M.S., M.B.A., and Ph.D. degrees from Ben-Gurion University in Israel.

1. Mamba Mentality: Success isn’t casual, it’s all in. We’re looking for people born with two settings: full throttle and sleep.

2. Focus Forward: There’s no point in getting worked up about things you can’t control. Use your energy to build a better product, write a winning proposal, or talk to a customer about their issues. Of course this principle is named after Ben Horowitz’s classic short essay. “All the mental energy that you use to elaborate your misery would be far better used trying to find the one, seemingly impossible way out of your current mess. It’s best to spend zero time on what you could have done and all of your time on what you might do. Because in the end, nobody cares, just run your company.”

3. Customer Obsession: Jeff Bezos famously banned slides at Amazon because that instrument is about persuasion, not truth. In lieu of slides, Amazon adopted a “Working Backwards” process which guides product strategy from the customer perspective, not from the typical internal/company perspective or competitor perspective. I’ll admit it. It felt really corny at first, as a three-person startup, to use the working backwards process for a new product, but it worked.

4. Clear Writing is Clear Thinking: We view the working backwards process as part of a broader writing culture at Exia. Having a writing culture is great for a remote team that works asynchronously. On a more personal level, I hate debating the same topic twice, and startups move so fast that people tend to forget things if they are not written down. It may sound boring, but we always have agendas, notes, decision documents, and monthly investor emails at Exia.

5. Intrinsically Motivated: Everything goes wrong at a startup. We’re looking for self-motivated people who will have the grit and animalistic hunger to figure out how to solve problems and grind out solutions.

We’re only a few months into our journey. As our business grows and evolves, our culture (and this document) will, too. Follow our blog for more updates on Exia Labs’ tech efforts, company culture, product developments and more.