Founding Engineer

[LegalOS](https://www.legalos.ai/) is building the **AI-native immigration law firm** for the world’s top talent and the companies that hire them.

High-stakes immigration cases—O-1s, EB-1s, L-1s, H-1Bs, and other employment-based visas—still rely on slow, expensive, manual workflows. LegalOS combines AI-powered document analysis, drafting, and workflow automation with licensed attorney oversight to prepare USCIS-ready petitions dramatically faster.

We know this market firsthand. Matthew, our CEO, previously founded Legal Bullet, an legal software company recognized by Forbes 30 Under 30. Matthew and Rachel, our COO, are siblings who grew up around immigration law through their father’s 40+ year practice. Claire, our CTO, brings engineering experience building AI and compliance systems at scale.

We’re turning that experience into software that helps companies hire and retain global talent—and gives clients and attorneys a better way to navigate the process.

We are currently a team of 8 based in San Francisco and are looking for our first 3rd-4th engineering hires to help us scale to our first 100 enterprise clients.

About the role

We’re hiring a **Founding Engineer** to build across the LegalOS product: client experiences, internal tools, backend services, and AI agents & workflows.

There is a lot of everyday full-stack engineering to do. You’ll ship features, fix bugs, improve existing flows, and work directly with the people using what you build. You’ll take work from an ambiguous request through design, implementation, testing, and release.

**Agent systems will also be a substantial part of the role.** You’ll help build the software that lets agents use tools, retrieve relevant information, work with documents and files, execute tasks in sandboxes, and carry out longer workflows reliably.

You’ll work closely with Claire and Matthew & our engineering team across our **Next.js/TypeScript application, Supabase/Postgres backend, AI workflows, document generation, and internal review tools**. You’ll also work with our team, on data and evaluations, and collaborate directly with attorneys and operations.

What you’ll do

* **Ship full-stack product work.** Build and improve intake, evidence collection, case dashboards, document uploads, notifications, and other everyday client and internal workflows.
* **Make the existing product better.** Investigate bugs, resolve reliability and performance issues, and improve usability across the application.
* **Work directly with users.** Understand how attorneys and operations actually handle cases, clarify requirements, and turn manual processes into practical software.
* **Develop agent capabilities.** Work on tool use, retrieval, orchestration, state management, and the supporting code that makes agents useful in production.
* **Build agent workspaces and execution environments.** Help agents read, create, and organize files, use sandboxes, preserve their work, and recover from failures.
* **Build AI-assisted document workflows.** Improve document processing, evidence mapping, case drafting, follow-up questions, and document generation.
* **Make outputs easier to review.** Build clear source attribution, version history, review interfaces, and human checkpoints so attorneys can verify and improve AI-generated work.
* **Measure and improve reliability.** Work with legal team on evaluations, use failures to guide improvements, and add appropriate tests and observability to the systems you ship.
* **Bring sound technical judgment.** Propose approaches, explain tradeoffs, and work with Claire and Matthew to agree on scope and technical decisions.

What success looks like in your first 90 days

**First 30 days:** Learn the product, codebase, case workflows, and attorney review process. Ship useful features and fixes, get comfortable working across the stack, and understand where clients and the team encounter friction.

**By 60 days:** Take a meaningful workflow from problem definition to release. Work directly with operations or attorneys, make sensible implementation tradeoffs, and follow through on feedback after shipping.

**By 90 days:** Consistently deliver across product work and AI workflows with increasing independence. Your work should produce a measurable improvement in manual effort, reliability, review quality, or the client experience.

We’re looking for someone who quickly becomes a dependable builder across the product and contributes meaningfully to how we develop our AI systems.

What we’re looking for

* **3+ years building production web applications**, or equivalent evidence of exceptional product engineering ability.
* **Strong TypeScript, React, and full-stack fundamentals.** You can move comfortably between frontend, backend, APIs, databases, and internal tooling.
* **Strong frontend taste.** You care about clear interactions, thoughtful details, and making complicated workflows easy to use.
* **Practical systems judgment.** You can reason about data models, permissions, integrations, background jobs, and failure modes without overcomplicating the solution.
* **Independent execution.** You can turn an ambiguous request into a clear plan, ship it, and take responsibility for how it works in practice.
* **Clear communication and collaborative judgment.** You ask useful questions, explain tradeoffs, incorporate feedback, and raise problems early.
* **A strong appetite for learning.** You’re excited to develop deep fluency in agent systems and the immigration workflows our software supports.

Very strongly preferred: experience building agent systems

**Hands-on agent-systems experience is a major advantage, but it is not a hard requirement.** Strong full-stack engineers who demonstrate the fundamentals, curiosity, and ability to learn quickly are encouraged to apply.

We’re particularly interested in experience with:

* Agents that use tools, retrieve information through RAG, manage state, and execute multi-step workflows.
* Agent workspaces, filesystems, sandboxes, and secure execution of generated code or tool calls.
* Production AI features involving structured outputs, streaming, retries, observability, evaluations, and human review.
* Frameworks and protocols such as **Vercel AI SDK, Mastra, and MCP**, or comparable tools.

We care about what you’ve built, the decisions you made, and what you learned when it failed.

Bonus points

* Experience with document-heavy products or complex operational workflows.
* Work in legal tech, fintech, healthcare, HR tech, compliance, or another field where correctness and sensitive data matter.
* Early-stage startup experience, substantial side projects, or meaningful open-source contributions.

This is a **high-ownership founding engineering role**. You should be comfortable talking to users, attorneys, and operators; understanding messy workflows; and building the product systems that make them 10x better.

If you want to help rebuild one of the most important legal infrastructure markets from first principles, we’d love to talk.