Working at Chariot Defense: Culture, Pace and Who Thrives

How Work Gets Done

Chariot Defense put a hybrid power system in the hands of the 2nd Brigade, 101st Airborne at the Joint Readiness Training Center six months after the company was founded. The company did not exist when planning for that exercise kicked off. By the time the brigade arrived, Chariot's system was in their trucks.

Founded in late 2024 and based in South San Francisco, Chariot operates with roughly 30 people across hardware, firmware, power electronics, and software, including alumni of Anduril, Tesla, Apple, Uber, and Archer, plus frontline military veterans. Their first product, the M424 system now branded as the Amphora line, moved from concept to JRTC in under half a year. That speed comes from a decision-making structure that collapses the distance between engineering and the user. CEO Adam Warmoth, who led counter-UAS at Anduril before founding Chariot, said the team builds "in a way that's informed by our experience getting into the mud and dirt with the warfighter." Engineers don't wait for requirements documents. They deploy with units, watch where the coffee pot trips the breaker and takes down the air defense radar, then rewrite the power-management firmware to stagger compressor starts and cut peak demand by a factor of three. The software layer that mediates bi-directional power flows, deciding when to draw from a vehicle alternator, a generator, solar, or shore power, exists because operators said, "we can't be PhDs in power."

Authority sits with the technical leads who own the subsystem end to end. A firmware engineer owns the microcontroller stack from bring-up through field validation. A power electronics engineer owns the silicon-carbide converter design through thermal qualification. No separate systems-engineering group writes interface control documents for someone else to implement. The same people who simulate transient response in the lab watch the telemetry when a reconnaissance company runs 36 hours silent at JRTC.

The workflow follows a tight loop: forward-deployed engineers embed with units, identify the signature or weight or reliability failure, return to South San Francisco, iterate hardware or software, and redeploy. The DIU awarded a $7.6 million, The Debrief's data shows, 12-month PRISM contract in September 2026 to validate the Amphora system at larger scales: battalion, then division, then corps. Warmoth told The Debrief that soldiers will measure power performance and fuel use in real scenarios, and that feedback will drive adaptation throughout the testing phase. Environmental and durability tests run in parallel.

This is not a culture of "move fast and break things." It is a culture of "move fast because the thing you're building keeps a command post from lighting up a thermal imager." The pace is set by the exercise calendar and the threat environment, not by a sprint board. When the Army's transformation guidance shifted from process compliance to outcome focus, Chariot was already operating that way because the only way to field a hybrid power system in six months is to treat the process as a tool, not the goal. Whether those principles survive scaling is the open question the next section addresses.

The company has raised $8 million in seed funding, PRNewswire reported, from top defense technology investors, led by General Catalyst and XYZ, with participation from Cubit Capital, Ravelin, Forward Deployed VC, Pax, New Vista, D3, and Brave Capital.

"Chariot Defense embodies the values-driven, mission-aligned companies that we seek to support, building a critical infrastructure layer that directly strengthens national security," says Paul Kwan, Managing Director at General Catalyst. "Their ability to go from concept to fielded systems in under six months, work alongside operators, and solve real-world challenges at the edge has been remarkable. We are excited to support Chariot as they build what we believe to be a foundational platform to power next-gen defense innovation."

Today's military still powers its ever-expanding suites of sensors, drones, autonomous systems and AI-driven command posts using noisy generators based on decades-old technologies that can't handle the demand. Chariot's technology bridges that gap.

"While everyone focuses on building smarter weapons, Chariot is building the intelligent power systems that makes them all work," says Ross Fubini, Managing Partner at XYZ Venture Capital. "Chariot sees this more clearly than anyone and is building the distributed power infrastructure that will define how future conflicts are fought. This isn't just better batteries or quieter generators—it's positioning to become the power prime contractor for modern warfare."

A New Infrastructure Layer for Modern War

Today's warfighters are increasingly reliant on high-tech systems, from autonomous drones and mobile command posts to advanced sensors and Counter-UAS platforms. But these systems are only as effective as the power sources behind them. Modern military operations require scalable, modular power, ensuring continuous, adaptable operation for forces in contact.

Built specifically for the needs of the modern warfighter, Chariot Defense's Amphora power systems extend operational independence for resilience during supply disruption, adapt to unit needs during distributed, expeditionary missions, and sustain rugged, quiet, all-weather operations. Amphora is silent, lightweight, and modular, providing reliable power with minimal signature, addressing major gaps in legacy generator-dominated systems.

Driven by Defense Experts, Informed by the End User

Chariot Defense's leadership team brings deep experience from Anduril, Tesla, Apple, Uber, Archer, and frontline military veterans. Warmoth brings a background in engineering at Stanford, problem and customer expertise from Anduril, and technical knowledge from cutting-edge aerospace companies to deliver much-needed disruption to the military power market. Their approach blends commercial agility with a defense-first mindset, rapidly fielding, iterating, and refining solutions based on real-world feedback.

"We're building for real-world needs, side-by-side with the people who will use and benefit from our platform," says Warmoth. "Our goal isn't just to provide better power systems—it's to become the power backbone that enables a new class of warfighting capabilities and makes next-generation military operations possible."

The Operating Principles

Chariot Defense doesn't publish a values page. Its operating principles show up in what the team builds, how fast it ships, and which problems it refuses to accept as constraints. The through-line across founder interviews, soldier feedback, and hiring signals is a cluster of commitments that function as de facto values: outcome over process, commercial speed on defense timelines, the signature reduction invariant, software-defined interoperability, and domestic supply-chain sovereignty. Each carries a trade-off the company has made explicitly.

Outcome over process is the clearest. Warmoth describes the Pentagon's new acquisition mindset as "our goal is to actually win and then you field the team that's going to win and you work backwards." Chariot's own path — DIU PRISM contract, Arctic delivery in 12 months, field testing with 4th Infantry Division at Fort Irwin — mirrors that logic. The trade-off: you don't get the luxury of a two-year requirements phase. Engineers ship hardware that soldiers use in weeks, then iterate based on what breaks. The company's LinkedIn posts frame this as "transforming in contact": flood the zone, learn fast, adjust. That means accepting field failures as data, not disasters.

Commercial technology transfer replaces the old defense R&D model. Warmoth's career, including Anduril, Archer, and Uber Elevate, and the team's pedigree (Tesla, Joby, Impulse) embody the flip: "Technology invented commercially, brought to department... we're kind of part of that trend." The Amphora architecture leans on silicon-carbide power electronics, high-voltage battery management, and software-defined conversion, all matured in EVs and eVTOLs. The trade-off: you inherit commercial supply chains and talent pools, but you must ruggedize for vibration, temperature, and EMI in ways no car maker worries about. Chariot's hiring, including power electronics, embedded Linux, firmware, and mechanical, reflects that dual mandate.

Signature reduction as a design invariant appears in every product spec and soldier quote. "Low thermal output, low acoustic output. So can't be hot, can't be loud, and it's got to be easy to actually move." The M424 let a multi-functional reconnaissance company run radios, EW gear, and drones for 36 hours with no detectable signature. At a battalion command post the same system delivers two to three hours, a deliberate sizing choice, not a limitation. The trade-off: energy density and thermal management drive every mechanical decision. You don't add a fan because it's easier; you design the enclosure to conduct heat because a fan is a signature.

Software-defined interoperability turns power into a protocol. "Someone will go plug in a coffee pot and it'll take down the air defense radar": Warmoth's example of today's dumb grid. Amphora inserts a "smart power layer" that negotiates voltage, frequency, and priority through "hot software handshakes." The company's coalition messaging makes this strategic: "Mismatched power standards pose not only a tactical challenge but also strategic vulnerability." The trade-off: you own the software stack from cell-level BMS to grid-level orchestration. That's a hiring burden, requiring firmware engineers who understand MIL-STD-1399 and CAN bus and real-time Linux, but it's the only way to make a Chinese-made power bank irrelevant.

Domestic supply-chain sovereignty is the newest principle, articulated in a September 2026 op-ed by Head of Government Relations Ben Kesling: the U.S. bans Chinese drones but fields Chinese power banks with Bluetooth and internet connectivity. Chariot's response is implicit in its bill of materials: final assembly in South San Francisco. The trade-off: higher unit cost and longer lead times than a Shenzhen integrator. The company bets the DoD's new risk tolerance, "willingness to accept risk... willingness to kind of bring in new entrants," will pay that premium for assurance.

Field utility over spec-sheet performance appears in unvarnished soldier feedback: "It has been giving us all the power that we need," "If we had two, we would have no problems," "Send the drones home, this is the most useful thing out here." The 45 percent fuel-reduction claim on the website, Chariot Defense's figures show, is a derived outcome, not a design target. The design target is "power precisely how and where it's needed": flexible, adaptable, controlled, drawing from any source. The trade-off: you build for the worst-case edge case (Arctic cold, contested logistics, coalition plug-and-play) and accept overcapacity in the common case.

Ownership of hard problems is the recruiting pitch: "If you want to own hard problems and see your work deployed in real conditions within weeks, come build with us." The job board bears this out: power electronics, embedded Linux, firmware, mechanical, manufacturing technician roles all in the same $130k–$180k band, all in San Bruno. No remote, no "staff engineer" titles. The trade-off: you hire generalists who can cross disciplines, not specialists who optimize one layer. The VP of Engineering hire, Satyan Chandra from Tesla, signals the bar: someone who has shipped high-voltage hardware at volume and speed.

These principles aren't posters on a wall. They're visible in the contract vehicle (DIU PRISM, not a traditional program of record), the deployment tempo (Arctic in 12 months), the org chart (VP Engineering hired early), and the product architecture (software-defined, coalition-ready, domestically assembled). The company is small (roughly 30 people), so the principles are still personal. The hiring bar that enforces them is the subject of the next section.

The Hiring Bar

Chariot Defense sits at the intersection of two talent pools that rarely overlap: the commercial electro-industrial sector, including high-voltage batteries, silicon-carbide power electronics, and software-defined power conversion, and the national-security apparatus that buys, fields, and sustains kit in combat. The hiring bar selects for people who can operate in both without translation loss.

The company's own framing makes the model explicit. Leadership cites Anduril (autonomy from self-driving cars) and Palantir (big-data infrastructure from cloud compute) as the template, echoing that same trend. That means the strongest signals on a resume aren't defense primes; they're Tesla and the power-electronics groups at companies pushing high-voltage hardware. The open roles bear this out:

Technical depth is necessary but not sufficient. The a16z Deep Dives interview repeats a single phrase: "outcome focus." The Army's transformation initiative let Chariot into a JRTC exercise six months after first check: no six-month paperwork cycle, no "try again next year." The company's job is to follow that same logic. Candidates who have shipped product in regulated, high-reliability environments signal they can move at that speed without breaking the certification chain.

Equally weighted is the ability to derive requirements from the user, not the spec sheet. Leadership describes the core task as understanding "from soldiers in the mud, cold, wet, tired, hungry, what makes their lives easier or better." That filters for engineers who have done field trials, supported customer deployments, or worked in applications engineering, roles where the feedback loop is measured in hours of radio silence, not Jira tickets.

Communication shows up in every first-party signal. This isn't culture fit as vibe check: it's a filter for people who can brief a program manager at 0600 and a four-star at 1400 without changing vocabulary.

The DoD baseline then applies a hard gate: U.S. citizenship, security-clearance eligibility (clean background, stable finances, limited foreign contacts), and tolerance for structured, hierarchical processes. An active Secret or TS/SCI cuts six to eighteen months off the start clock. The company doesn't advertise this, as it's the price of admission for any firm playing in Army transformation, but every hire knows it or learns it fast.

What emerges is a narrow profile: commercial power-electronics or embedded-software depth, proven in regulated shipping environments, with demonstrated user-facing problem discovery, fluent technical communication, and a clean enough life to hold a clearance. The bar doesn't select for defense pedigree. It selects for engineers who can make commercial tech survive the mud. But passing the bar is only half the story; staying is the other half.

What the Reviews Don't Tell You

Public employee reviews for Chariot Defense are nearly nonexistent, unsurprising for a company founded in 2024 with roughly 30 people. Glassdoor shows 20 reviews for "PACE, Inc." and a separate listing for "Intensity," but neither maps cleanly to Chariot Defense's current branding or entity. The company's own LinkedIn page lists 5,483 followers as of late 2024, a figure that reflects recruiting reach more than employee sentiment. Zero G Talent's board data shows active postings for the six core engineering roles in San Bruno, spanning embedded Linux, firmware, power electronics, electronic design, mechanical, and systems, all banded at $130,000–$180,000, plus a separate cluster of business roles under "Chariot" in New York City ranging from $190,000 to $350,000. Whether the NYC postings belong to the same entity or a different company sharing the name is unclear from the data.

The absence of review volume is itself a signal. At this stage, anyone joining is writing the culture in real time, not inheriting it. The leadership pedigree, including Anduril, Tesla, Apple, Uber, Archer, and military veterans, sets a baseline expectation for technical rigor and pace, but no public forum has yet captured how that translates to daily experience. The $7.6 million DIU award for the Amphora power system, announced by The Debrief in September 2026, confirms external validation of the technical roadmap, yet employee perspectives on workload, autonomy, or mission alignment remain undocumented in aggregate sources.

For a candidate, this means due diligence shifts from reading reviews to direct conversation. Ask the hiring manager how decisions get made when the founding team is in the room. Ask a peer interviewer what surprised them after three months. Ask what the team celebrated last quarter and what they argued about. The answers will be more reliable than any Glassdoor aggregate because the aggregate doesn't exist yet. The pattern of who lasts and who leaves, visible in the company's operating tempo, fills the gap.

Who Lasts, Who Leaves

The profile of someone who lasts at Chariot Defense looks less like a traditional defense contractor employee and more like a special-operations technician who can write embedded C. The company's small team operates at the intersection of three unforgiving domains: high-voltage power electronics, DoD acquisition reform, and expeditionary field testing. Every hire sits in the overlap.

People who thrive share a cluster of traits visible in the company's operating pattern. First, they tolerate — even prefer — ambiguity with a deadline. The a16z Deep Dives interview describes a "transforming in contact" model: flood the zone with capability, learn what breaks, iterate. Six months from first check to a field exercise at Fort Irwin. Twelve months from DIU PRISM award to Arctic delivery. That cadence selects for engineers who can ship hardware without a mature requirements document, then debug it in the cold while a soldier watches. The "outcome focus" the founders cite as their central tenet means the process serves the result, not the reverse. If you need a spec freeze before you start layout, you will miss the window.

Second, they cross discipline boundaries without friction. Chariot's product — Amphora — is simultaneously a battery pack, a bidirectional inverter, a thermal-management system, and a software-defined power router. The job board lists those same core engineering roles all in the same $130–180k band. In a small company, those roles bleed into each other. The firmware engineer who refuses to look at a schematic because "that's hardware" becomes a bottleneck. The mechanical engineer who treats the PCB as a black box forces a redesign. The people who stay are the ones who walk to the bench, probe the rail, and suggest the firmware fix that saves a respin.

Third, they treat the warfighter as the direct customer, not the program office. The company's own language is explicit: "Our product is driven by comprehensive live end-user engagement." The Fort Irwin testimonial quotes a soldier: "This piece of equipment increases the sustainability of a platoon by days." The Arctic deployment puts systems in front of troops early in the contract period. Engineers who have never spoken to a user, or who view "requirements" as a contract artifact rather than a conversation with a 19-year-old operator in a Bradley, will build the wrong thing. The company's hiring nights with Beacon AI (SF Tech Week, September 2026) pitch "design that system, build it, and deploy it to the people who depend on it." That loop — design, build, deploy, watch, fix — is the job.

Fourth, they accept regulatory and security friction as terrain, not a blocker. That op-ed frames Chinese-manufactured power banks as a cybersecurity and explosive vulnerability. That argument, covering technical, supply-chain, and policy aspects, is the environment Chariot sells into. Employees who treat ITAR, CMMC, and classification as someone else's problem slow the team. The board — Gen. (Ret.) James Rainey, Ellen Lord, Stephen Rodriguez, Brig. Gen. (Ret.) Samuel Peterson — signals that the company navigates at the flag-officer level. Junior staff still have to execute the paperwork that makes those relationships usable.

Who burns out? The pattern inverts cleanly. People who need process maturity, such as defined roles, stable sprints, and a QA gate before hardware ships, will find the pace chaotic rather than liberating. The "willingness to accept risk" the founders praise means the company ships prototypes to Army brigades on 12-month timelines. If your risk calibration requires a design review board, you will be the bottleneck.

People who optimize for local technical elegance over mission utility struggle. The a16z interview recounts a software tweak that cut peak power demand by 3×, shrinking the whole system. That kind of cross-layer insight, such as firmware changing mechanical weight, is rewarded. A firmware engineer who optimizes interrupt latency for its own sake while the thermal margin evaporates is not.

People who cannot context-switch between the bench and the briefing room will stall. The same week might demand a thermal-chamber validation, a DIU program review, and a conversation with a SOCOM operator about why their EW suite browned out when the coffee pot kicked on. The company's scope, including national security, defense technology, power systems, operational energy, air defense, communications, ISR, UAS, unmanned systems, electronic warfare, directed energy, hybrid powertrains, embedded software, electric vehicles, and tactical power, is not a keyword list. It is the daily vocabulary.

People who treat the mission as branding rather than a constraint will leave or be pushed out. The DIU PRISM contract ($7.6M, 12 months, Arctic delivery) is a real deadline with real consequences. The board of advisors includes the former Under Secretary of Defense for Acquisition and the former Commanding General of Army Futures Command. They are not ornamental. The company's survival depends on converting field demonstrations into programs of record. That pressure is not performative.

The research offers no Glassdoor corpus, no Blind threads with substance, no longitudinal attrition data. The company is too young (founded 2024) and too small. But this pattern — rapid fielding, cross-disciplinary hardware/software, direct warfighter feedback, senior DoD sponsorship, outcome-over-process rhetoric backed by 6-month exercise cycles — creates a clear filter. You thrive if you are a systems thinker who builds hardware, writes code, talks to soldiers, and treats bureaucracy as a puzzle to solve en route to the Arctic. You burn out if you need any one of those to be someone else's job. The next exercise is already on the calendar.

Working in frontier tech? Zero G Talent tracks the openings: see every open Chariot Defense role, browse frontier tech jobs, openings at Chariot, and the people building the field.