Unknown Company
Applied AI Engineer - Internal at Unknown Company in San Francisco, CA
Skills
Job Description
About Matter Intelligence
Matter Intelligence is building the future of vision AI: pairing a world-first sensor that sees molecular chemistry, temperature, and 3D shape with a Large World Model—the most powerful intelligence engine for the physical world. This system doesn't just see what something looks like; it understands everything from a single pixel. We call this Superintelligent Vision.
Our team has delivered technologies to Mars for NASA/JPL, co-founded and led infrastructure for OpenAI, designed cutting-edge sensors for U.S. Defense, and invented core algorithms for spectral and 3D imaging. We've come together to build the next infrastructure for vision and intelligence in the physical world.
About the Role
Matter is building technology at the intersection of hardware, AI, and Earth observation. This role sits at the boundary of Engineering and Operations.
As the first hire in this role, you'll work directly with teams across Matter: sensor engineers, ML researchers, satellite systems designers, sales, and leadership. Your job is to understand how they work, identify where intelligent agents can compress hours into seconds, and build them. You might spend a week designing a multi-agent workflow to automate competitive intelligence tracking, then shift to an internal tool that gives the hardware team instant retrieval across five documentation systems. You own problems end-to-end - from discovery through deployment through iteration.
This isn't a role for someone who wants to follow a ticket queue. It's for a developer with strong CS fundamentals who can operate with autonomy, earn trust across a demanding organization, and ship production-grade agents that people actually depend on.
Key Responsibilities
Agent Development & Engineering
Design, build, test, and maintain production-grade AI agents and tools using current frameworks and APIs.
Architect agentic systems that are reliable, secure, observable, and maintainable.
Build AI-native UX patterns that reduce friction to adoption: human-in-the-loop checkpoints, auditability, graceful failure/rollback, and feedback mechanisms.
Select and integrate appropriate tools, memory systems, and retrieval strategies for each agent's use case.
Write clean, well-documented, version-controlled code. You set the standard.
Product Discovery & Requirements
Interview stakeholders across engineering, hardware, science, sales, and leadership to identify high-leverage opportunities for automation and value creation.
Run user interviews, workflow audits, and surveys to understand actual bottlenecks before writing a line of code.
Translate qualitative findings into clear, scoped requirements and build plans.
Maintain a prioritized backlog of agent opportunities with transparent reasoning on what gets built next and why.
Cross-Functional Partnership
Work across every part of Matter; no two weeks will look the same. One week you're building a pipeline tool for the payload team; the next you're automating how sales tracks relationships.
Be responsive, communicate clearly, and follow through - your coworkers are your clients.
Own rollout and change management for what you build: onboarding, adoption, feedback loops, and iteration based on real usage.
Share and document best practices for working with AI tools across the organization. Lead training session on the best practices for AI adoption and use.
Quality, Security & Documentation
Build with security first: apply proper secret management, access control, and prompt injection defenses from day one.
Write agent documentation that humans can actually use—runbooks, decision logs, API specs.
Establish evaluation frameworks and monitoring so stakeholders know when an agent is working and when it isn't.
Own quality end-to-end: you ship it, you support it, you improve it.
The Tooling Landscape You Should Know
AI agent development moves fast. We expect fluency across the current stack (examples, constantly changing):
Agent frameworks & orchestration: Anthropic SDK / Claude API, MCP, Skills, orchestration tools
Workflow/process automation: n8n, Zapier/Make, Retool workflows, Airtable automations (or equivalents)
AI-native development: Claude Code, Codex, Cursor, GitHub Copilot
Observability & evaluation: LangSmith, Langfuse, Braintrust, DeepEval (or equivalents) - you instrument what you build and measure whether it works
Data & memory: Vector databases (Pinecone, Chroma, Weaviate, pg-vector), RAG pipeline design, structured vs. unstructured retrieval tradeoffs
Security: Secret management (AWS Secrets Manager, HashiCorp Vault), OAuth 2.0, RBAC, least-privilege design, prompt injection defense
Documentation & API design: Structured logging, clear runbooks - because agents that nobody understands get turned off
Qualifications
Required
B.S. in Computer Science or equivalent practical experience.
Demonstrated ability to build and ship production software: clean code, version control, testing, CI/CD.
Hands-on experience building with LLM APIs and/or agent frameworks — you've built something real, not just followed a tutorial.
Strong product instincts: you can sit with a stakeholder, understand their workflow, identify what actually matters, and scope something achievable.
Security-conscious by default: you think about access control, data handling, and failure modes before they become problems.
Clear written and verbal communicator — you can write documentation a non-engineer will actually read.
Preferred
Experience building internal tools or automation for operational teams (BizOps, RevOps, FinanceOps, etc.).
Familiarity with agentic design patterns: tool use, memory, multi-agent coordination, human-in-the-loop flows.
Experience with agent evaluation and observability tooling (LangSmith, Langfuse, Braintrust, or similar).
Comfort with RAG architectures and vector database selection.
Prior experience in a startup environment where you had to own end-to-end delivery without heavy infrastructure.
What Success Looks Like
Within 60 days, you've shipped at least one agent or automation that a team at Matter depends on.
Your stakeholders can tell you exactly what problem you solved for them and why it worked.
The agents you build are instrumented, documented, and maintainable by someone other than you.
You have a clear view of the next 3–5 highest-leverage opportunities across the company and a defensible case for what gets built next.
People across Matter know your name and trust that you'll follow through.
Location
This role is based in San Francisco, CA, with onsite presence required.
ITAR Requirements
To comply with U.S. export regulations, applicants must be one of the following:
A U.S. citizen or national
A lawful permanent resident (green card holder)
Eligible to obtain required authorizations from the U.S. Department of State
Employee Offerings & Benefits
Compensation: Competitive total package based on experience.
Equity: Early-stage equity package so you share directly in Matter's growth and success.
Health & Wellness: 100% employer-paid health, dental, and vision coverage.
Growth: Unique exposure across hardware, software, science, and commercial teams—and a front-row seat to building the world's most capable sensing and intelligence platform.