Founding Engineer - AI

PayOS | San Francisco | $130k – $170k + Significant Equity

The Mission
AI agents are ready to run the economy, and PayOS is readying the financial infrastructure for it.

The Opportunity
We are looking for a Founding AI Engineer.
We need you to own the end-to-end Agentic Layer for PayOS. This is a role for a "10x" builder with 3+ years of experience who wants to create a difference in commerce & fintech for the agentic era.

• Are you entrepreneurial and/or have you ventured on your own before?
• Do you feel like you have it in you to be a General Manager of what you will build - build, sell, support?
• And when you build, you are obsessed over UX?
• Are you the AI person who knows what’s the next big thing the world needs?

If Yes, please apply immediately.

Your Stack & Responsibilities

• Live Inside the New Stack: We maximize leverage by embracing AI. You don’t just write code; you generate it using Claude, Gemini, ChatGPT and other models/tools. You leverage long-context windows and prompt caching to amplify your output far beyond what is typically possible for a single person.
• MCP Native: You didn't just read the announcement; you have already built an MCP server. You understand how to standardize context between models and tools.
• Agent Architecture: You will design and build autonomous agents that execute multi-step tasks (not just chat). You already have experience building agents, managing tool-calling, and implementing MCP (Model Context Protocol) servers. You have battle scars from LangChain, CrewAI, or custom loops.
• Full Stack AI (TypeScript + Python): You are fluent in Python for the agents, but equally dangerous in TypeScript/React/Next.js for the application layer. You can build the "Generative UI" to make the agent's work visible and beautiful.
• Memory & Context Strategy: You have already built long-term memory systems for agents and dealt with the issues that follow retrieval drift, context pollution, and "lost-in-the-middle" problems.
• Evals & Reliability: In Fintech, 90% accuracy isn't enough. You know how to build evals to systematically test agents, reduce hallucinations, and ensure reliability before money moves.
• Forward-Deployed Engineering: You will frequently partner with customers to integrate our platform. You will build custom demos, "Best of the Web" agents, and workflows that prove the value of PayOS.

Our DNA (Who You Are)

Passion & Agility (High Intensity): We move incredibly fast. We are a team that does "whatever it takes" to move the needle. This often means working with high intensity and being responsive outside of traditional hours because we are deeply invested in our mission.

Concrete Outcomes > Activity: We don't care about "effort"; we care about solving our customer’s problems and building the services they need. You measure your success by the concrete value you bring to the company every day. You ship code that works, not just code that looks good.

Deep Product & Market Fluency: You must go beyond the code. You need to deeply understand the products we have and exactly what the market demands. Success here means identifying customer gaps and architecting solutions that resonate immediately.

*Ownership & Autonomy: *
We don’t have playbooks. You are a self-learner who spots a problem and solves it immediately without waiting for instructions. You roll up your sleeves and wear multiple hats.

One Team, Zero Politics: Because we are small, chemistry is everything. We prioritize total transparency. If you face a roadblock, you share it early. We view challenges collectively: an issue you face is an issue for all of us.

Benefits & Perks

• Top-Tier Healthcare: 100% Employee-Covered Health, Dental, and Vision insurance.
• Meaningful Equity: You are joining the Founding Team.
• Direct Mentorship: Work directly with a world-class engineering team.
• Cutting Edge: Work on the absolute frontier of Fintech x AI.

How to Apply
We do not want a standard cover letter. Please send an email to [email protected] with:

1. Your GitHub/Portfolio.
2. A link to a LIVE AI project you have built (deployments > tutorials).
3. The "Bug Report": A 2-3 sentence description of the hardest, weirdest debugging challenge you faced with an LLM (e.g., fixing a memory loop, reducing hallucination).