Product Engineer

Our client is an open-source AI gateway gives developers and enterprises one reliable, provider-agnostic way to build across OpenAI, Anthropic, Gemini, Bedrock, self-hosted models, and emerging agent tooling.

Rather than rebuilding integrations every time the model landscape changes, developers can build once on the platform and retain the flexibility to route, govern, observe, and scale AI traffic across providers. The company is approximately 20 people, at roughly $22M ARR, and has grown ARR 6x over the past year.

You will work closely with founders, engineers, and the developer community to make the gateway easier to use, more reliable in production, and more extensible as the AI ecosystem evolves. The work is not model-training research, and it is not a narrow ticket queue. The best solutions will come from understanding what a developer is actually trying to accomplish, then building the cleanest abstraction that gets them there.

You will have meaningful ownership over parts of the open-source product surface, from the initial user signal through design, implementation, release, and iteration. Areas of work will include:

• Building the interoperability layer that lets developers use many LLM providers through a consistent API.
• Designing request and response transformations, streaming behavior, tool calling, error handling, retries, and provider-specific capabilities.
• Creating provider-agnostic gateway features for emerging AI workflows, including MCP, agents, skills, compaction, batching, session management, and authentication.
• Improving the developer experience across the Python SDK, APIs, documentation, examples, and defaults used by a large global community.
• Diagnosing challenging production problems involving performance, reliability, compatibility, and developer workflows.
• Turning feedback from issues, pull requests, and user conversations into thoughtful product decisions rather than treating each request as a literal specification.

• Strong software-engineering fundamentals and a track record of owning production-quality systems.
• Strong Python skills, or clear evidence that you can become effective quickly in a Python-heavy codebase. Experience with FastAPI, APIs/SDKs, PostgreSQL, Redis, distributed systems, or developer infrastructure is helpful.
• Genuine curiosity about the AI application stack: LLM APIs, agentic systems, MCP, tool use, model-provider differences, and the messy edge cases that emerge in production.
• Evidence of initiative beyond a narrow job description, such as startup or founding-engineer experience, meaningful side projects, open-source work, hackathons, technical communities, or leadership.
• Product instincts: you start with the user problem, know how to run a focused experiment, and care about the clarity and durability of the eventual solution.
• Clear, direct communication. You should be comfortable explaining a technical tradeoff to developers, customers, and teammates without unnecessary complexity.

This is an opportunity to work on the infrastructure layer beneath modern AI applications - the part that makes AI systems portable, dependable, and practical for developers building in the real world.

• Massive open-source leverage: Your work will be used directly by a large and fast-growing developer community.
• Fast feedback loops: You will work closely with users and founders, see the impact of what you ship, and influence the roadmap.
• Real autonomy: There is room to identify important problems, make decisions, and own solutions end to end.
• A builder culture: The team values ownership, customer care, grit, growth mindset, and simple communication.