Trase
An AI agent operating system targeting regulated industries — healthcare and defense — backed by a serial-founder team out of Red Cell Partners
Trase is an agentic AI infrastructure company whose development began in September 2024 in McLean, Virginia, launched as an internal incubation project of the venture studio Red Cell Partners. The company positions itself as an “AI agent operating system,” providing an orchestration, governance, and security layer that spans cloud, on-premises, and edge environments, with healthcare and defense — two of the most heavily regulated industries — selected as its initial target verticals.
Dropped out of Brown University at 21 to found digital health company Rally Health (originally Audax) in 2010. The company was acquired by UnitedHealth Group in 2017 in one of the largest exits in digital health at the time, after which Verstandig served as UnitedHealth Group’s Chief Digital Officer from 2017 to 2021. Around 2018 he founded Red Cell Partners, co-founding companies including defense-tech firm Epirus (which has raised more than $500M cumulatively), defense AI company DEFCON AI, and precision-medicine company Zephyr AI. Has served as an advisor to the National Security Agency (NSA).
A former U.S. Army infantry officer who subsequently built a finance career at Goldman Sachs and Susquehanna International Group (SIG). He later transitioned into engineering, working as a principal engineer at Reddit, Dropbox, and Google, where he led full-scale mobile app redesign and modernization efforts. Laws combines military decision-making structures with large-scale Big Tech systems-architecture experience to lead Trase’s multi-agent orchestration design.
The executive team also includes Dr. Baskar Sridharan as President, who previously served as Vice President of AI/ML services and infrastructure at AWS, where he led development of Bedrock and SageMaker. Prior to AWS, he spent 16 years at Microsoft working on cloud infrastructure products including Azure Data Lake and Cosmos DB. Sridharan has cited his firsthand experience watching AI prototypes at Big Tech struggle to make the leap into production-grade products as a key reason for joining Trase. Beyond the core leadership team, the company’s National Security Advisors include former U.S. Secretary of Defense Mark T. Esper and former NSA Deputy Director Veronica Daigle, meaning government-relations capability for securing defense customers is built into the organization from an early stage.
Trase’s core product is an agent-orchestration operating system called “Trase OS.” Rather than functioning as a single task-automation chatbot, it is designed as a governance layer that allows multiple AI agents to collaborate across cloud, on-premises, and edge environments while preserving data sovereignty. Co-founder Joe Laws has compared the architecture to white blood cells coordinating dynamically across the body, describing how multiple agents can exhibit collective intelligence with minimal configuration to drive efficiency gains.
| Customer / Area | Application | Stage | Notes |
|---|---|---|---|
| Duke Health | Hospital administrative workflows (prescription refills) | Live | Reported 3x faster prescription-refill processing and roughly $193K in annual savings (as of November 2025 stealth-exit disclosure) |
| U.S. Navy | Defense administrative/operational workflows | Contract in progress | Specific contract scope and value undisclosed. Defense procurement track differs from healthcare in validation timeline and process |
| Additional undisclosed customers | Healthcare/defense vertical expansion | Undisclosed | The company has not disclosed customer count or additional customer names |
| Expansion beyond regulated industries | Verticals outside healthcare/defense | Exploratory | Company plans to use the $107M Seed proceeds to explore less-regulated industries, and is reportedly considering self-service features that would let customers build their own agents |
Note on unverified claims: The reported 3x improvement in prescription-refill processing and the $193K in annual savings at Duke Health were disclosed by the company to the press (Technical.ly) at the time of its November 2025 stealth exit, and are not independently verified third-party figures. The claim of outperforming hyperscalers on the GAIA leaderboard is likewise a company-reported figure, and no cross-verification against a publicly posted benchmark ranking has been confirmed. Revenue scale, valuation, and the scope and value of the Navy contract all remain undisclosed.
Trase has raised approximately $117.5M cumulatively in a span of roughly seven months — from a $10.5M Pre-Seed disclosed alongside its November 2025 stealth exit, to a $107M Seed announced in June 2026. Raising a Seed round in excess of $100M ahead of a Series A is an unusually large amount by industry standards, and is widely interpreted as a signal of strong investor conviction in AI agent infrastructure for the high-stakes, heavily regulated healthcare and defense sectors.
Development began as an internal incubation project within Red Cell Partners. The team spent this period refining a proprietary orchestration and reinforcement-learning-based platform, operating on internal capital prior to any external fundraising.
Co-funded by Red Cell Partners and Virginia Venture Partners, a fund managed by the Virginia Innovation Partnership Corporation (VIPC), a state-backed innovation agency. The company formally launched out of stealth with former AWS AI/ML Vice President Baskar Sridharan joining as President, and disclosed its early customer relationships with Duke Health and the U.S. Navy.
Led by Arch Venture Partners, a firm best known for its life-sciences and biotech investing focus. Disclosed via an Axios Pro exclusive, with CEO Grant Verstandig confirming the round directly. At roughly 10x the size of the Pre-Seed, this is an unusually large round for a pre-Series A company.
Note on undisclosed information: As of this writing, Trase’s company valuation, the specific terms of its Navy contract, and the revenue split between its healthcare and defense business lines remain undisclosed. Axios Pro has reported that the upcoming Series A is being pursued at a “hundreds of millions of dollars” scale, but the specific target amount and timing remain unconfirmed.
The enterprise AI agent market already includes numerous adjacent competitors: Palantir in data analytics and orchestration, Innovaccer in healthcare data integration, and clinical-documentation specialists such as Abridge and Ambience. In the defense sector, traditional defense IT services firms — CACI, Booz Allen, RTX, and others — are also expanding AI-driven analytics offerings. Within this landscape, Trase’s structural differentiation can be observed across three layers: its business model, its founding-team network, and its design specialization for regulated industries.
Rather than upfront licensing, Trase charges in proportion to realized efficiency gains, lowering the bar for pilot approval among conservative healthcare institutions and federal procurement offices. This model is double-edged, however: at scale, proving attribution of efficiency gains in complex, multi-system environments could constrain the pace of revenue recognition.
Former Secretary of Defense Mark T. Esper and a former senior NSA official serve as National Security Advisors — a network and credibility base for defense procurement that is rare for a startup of this age. The Navy contract serves as early evidence that this network effect is materializing.
Red Cell’s track record incubating healthcare and defense companies — including Epirus, DEFCON AI, and Zephyr AI — applies equally to Trase. This gives the company a structurally favorable starting position relative to standalone startups in fundraising, talent acquisition, and early customer access.
President Baskar Sridharan, who led AWS Bedrock and SageMaker, and Chief Architect Joe Laws, who led mobile architecture work at Reddit, Dropbox, and Google, bring direct, hands-on experience building large-scale distributed systems in production environments.
The platform is built around a governance architecture that keeps data from leaving the customer’s environment — a core requirement for healthcare and defense buyers. This positions Trase as purpose-built for the compliance and audit demands of regulated industries, in contrast to general-purpose AI infrastructure.
The Duke Health case provided a concrete operational metric — 3x faster prescription-refill processing — from the moment of stealth exit. This figure is company-disclosed rather than independently verified, a point worth flagging for investment due diligence.
Implications of an Arch Venture Partners-Led Round: Arch, founded in 1986, has historically concentrated on biotechnology and life sciences, with a contrarian investment philosophy centered on betting on “disruptive science” rather than conventional enterprise software. A firm with this orientation leading the Seed round for a healthcare AI infrastructure company can be read as a signal that Arch views Trase less as a typical SaaS tool and more as a platform with the potential to reshape clinical and health-system workflows at a structural level. That said, Arch’s core expertise sits in clinical and life-sciences domains, so whether a comparable network effect will materialize on the defense side of Trase’s business remains a separate question requiring independent validation.
As of June 2026, Trase remains a private, pre-Series A company, and key financial metrics — revenue scale, customer count, and valuation — are not publicly disclosed. Total funding raised to date is approximately $117.5M, and the company has indicated plans to pursue a Series A round of $125M or more in the second half of 2026. This analysis relies on company disclosures and press reporting given the nature of a private-stage company, and any figures that have not been independently verified are flagged separately throughout this report.
Opportunities include: a Seed round exceeding $100M ahead of Series A, signaling strong market conviction; a performance-based pricing model that structurally lowers the decision-making barrier for conservative healthcare and defense buyers; the proven repeat company-building capability and national-security advisory network of Red Cell Partners; high-profile early reference customers in Duke Health and the U.S. Navy; and the strategic credibility implied by a lead investment from biotech powerhouse Arch Venture Partners in healthcare AI infrastructure.
Risks include: limited investment due-diligence visibility given the broad non-disclosure of core financial metrics (revenue, customer count, revenue-recognition pace); the potential for disputes over attribution of efficiency gains under the performance-based pricing model in complex, multi-system environments — a structural risk to the stability and scalability of revenue recognition; execution complexity from operating simultaneously across two distinct procurement and compliance tracks (healthcare and defense), particularly given that Arch’s core expertise is concentrated in healthcare, which may limit the credibility-enhancement effect on the defense track; direct competition with already well-capitalized, established players such as Palantir and Innovaccer; the fact that the disclosed Duke Health performance metrics are company-reported figures without independent verification; and the potential for management-resource dilution given that founder Grant Verstandig is concurrently involved with multiple other Red Cell portfolio companies (including Epirus, DEFCON AI, Zephyr AI, and MyPattern).