Founding / Principal Engineer, Trust & Control Infrastructure for AI Agents

Stealth · Westport, CT (NY Metro / Fairfield County) · Hybrid, 1–2 days a week in person to start · Q4 2026

Join the Founding Engineer Team at a founder-funded AI trust infrastructure company. Working prototype in review and test, development partners next, pre-seed/seed financing in progress, founder has gone founding-to-IPO before.

The problem

Agents are being handed real authority. They approve refunds, move records between systems, change production configuration, and negotiate with other agents. Today no one can prove what an agent saw, decided, or did. That record lives in a model vendor's logs or a framework's trace, if it exists at all, and watching an agent is not the same as verifying it.

What an agent is shown decides what it does, so an agent's context and decisions are an attack surface nobody owns a control point over. Humans and systems need to constrain, correct, and sign off on agent decisions, before and after the fact. None of this can depend on one framework or model: a fleet built on LangGraph, CrewAI, AutoGen, the OpenAI Agents SDK, the Anthropic Agent SDK, Google ADK, and MCP needs one independent standard underneath all of them.

We are building that standard, so any agentic deployment, from a team's first ten agents to an enterprise's ten thousand, gains control over and insight into what its agents do and reaches production with trust and confidence. How we do it stays under wraps until the development-partner phase.

What you will build

The framework integration layer: our product running with at least two of the frameworks above, with identical guarantees, within your first quarter.

The verification path: a design partner's security team can verify an agent's decisions end to end without us in the room.

The control surface: how a human or policy engine constrains, corrects, and signs off on an agent's decision, and how that holds when the agent is fast and the human is not.

The performance envelope: guarantees that cost agents milliseconds, not seconds. Measured, and published.

The production stack: a working prototype exists and is in review and test. You review it, test it, and design the production stack from it. Your choices of languages, services, testing, agent tooling, and on-call become how we work.

What you bring

Six or more years building distributed systems or data infrastructure in production; you have owned a service other teams depended on. In the last two years part of that has been agentic: tool-calling models, agent runtimes, or MCP servers that other engineers' agents consume.

Real time in at least one of: data infrastructure (databases, event streams, ledgers), security or integrity engineering, observability and audit systems, or developer platforms. Lineage from a company whose product was the record other systems trusted counts double.

Founding or pre-scale stage, or scaled a team from a handful to dozens.

A degree in computer science, engineering, mathematics, or physics, or graduate work, publications, or patents in distributed systems, databases, cryptography, or LLM systems: a plus, not a gate.

Must have

You have built a system of record: an append-only, event-sourced, or ledger-style store where history is immutable, replay is deterministic, and people outside your team relied on it as the truth.

Cryptographic integrity in practice: hashing, signing, content addressing, verification by a party who was not there, with key rotation and algorithm change designed in up front.

Data lifecycle under constraint: deletion or retention designed into a system that also had to stay verifiable (regulatory erasure, legal hold, audit).

A developer-facing surface you were accountable for: an SDK, library, API, or MCP server that other engineers and their agents build on, with versioning discipline and documentation both humans and models can read.

AI-native in production: you shipped something where a model or agent is a production component, wrote evals, handled non-determinism, built guardrails and human-in-the-loop approval, and dealt with prompt injection. Production experience with at least one of the frameworks above.

AI-native in practice: you build with coding agents (Claude Code, Codex, Cursor, or the like) under your direction and are judged on what ships.

Fluent in Python and TypeScript.

Strong signal

Open-source stewardship; standards and protocol work (MCP, A2A, OpenTelemetry GenAI); multi-tenant design for people and agents; performance engineering; shipping into regulated or audited environments; Go or Rust in production.

The founding-era bar

Moxie, intelligence, ambition, and signal awareness: you take the hard path when it is right, reason from first principles, want to set the standard an industry adopts, and say out loud which three things matter this week. If you would choose this over a lab offer, your three sentences should say why.

You can be in Westport one to two days a week to start, with the rest remote: already in Fairfield County or the New York metro.

How we work

Westport, Connecticut, in person one to two days a week to start, remote the rest. We are on the Metro-North New Haven Line, just over an hour from Grand Central. AI-native by default: we use agents to build the thing that governs agents. You would be one of the first three engineers, working directly with the founder, a repeat enterprise-software operator who has scaled teams from founding stage through IPO and acquisitions. Decisions in days, not quarters.

Compensation

Base salary: $170,000–$240,000. This is the range we have set in good faith for this position; where you land depends on demonstrated depth. Founding equity, sized to experience and scope, four-year vest, one-year cliff. Benefits: medical, dental, and vision coverage; 401(k); flexible paid time off.

How we hire

Step 1: 30-minute conversation with the founder.

Step 2: 60-minute technical deep dive on something you built that had to scale.

Step 3: A working session in Westport on a real problem, under NDA, or straight to step 4 when the fit is clear.

Step 4: References, then offer.

A working-together path: begin by meeting and ideating with us before either of us commits to the full-time seat. Say in your application which path you want.

How to apply

Email opportunity@stealthaitrust.ai with the subject "Founding Engineer — Westport": one artifact showing scale design or models-in-the-loop work (a repo, design doc, postmortem, or talk), three sentences on why this problem, and your LinkedIn profile saved as a PDF. No cover letter.

Notices

Pay transparency. The base salary range and benefits above are provided in accordance with Connecticut's pay transparency law, as amended by Public Act 26-12 (effective October 1, 2026). Equity and benefits are in addition to base salary.

Salary history. We will not ask about your current or prior compensation.

Fair chance. Consistent with Connecticut's fair chance employment law (Conn. Gen. Stat. § 31-51i), we do not ask about arrests, criminal charges, or convictions on an initial application, and we consider qualified applicants with records.

Equal opportunity. We are an equal opportunity employer. We do not discriminate on the basis of race, color, religious creed, age, sex, pregnancy, gender identity or expression, sexual orientation, marital status, national origin, ancestry, present or past history of mental, intellectual, learning, or physical disability, genetic information, military or veteran status, status as a victim of domestic violence, erased criminal history record information, or any other characteristic protected by federal, Connecticut, or local law.

Accommodation. If you need a reasonable accommodation at any point in the application or interview process, tell us and we will provide it.

Work authorization. You must be authorized to work in the United States. We can support transfers of existing H-1B, O-1, or TN status; we are not initiating new visa sponsorship at this time.

Employment is at-will.