Agentic AI Engineer

You will join the Applied AI team at 3Commas, a small product team building QuantPilot, an AI trading copilot that helps serious retail traders research markets, generate strategies, backtest them and iterate together in a chat. The central problem you'll solve is making agents produce statistically sound strategies, not just plausible-looking ones, and proving it with rigorous evaluation.

What you'll do

• Own the agent architecture end to end: orchestration, tool loops, context management, sandboxes and model routing, including the quality, cost and latency trade-offs of running it in production.
• Own the contracts between the engine and the product, including how the frontend renders agent events.
• Build and own the evaluation layer: offline evals, backtest-based quality metrics and regression detection for every prompt or model change.
• Bring statistical rigor to strategy evaluation through overfitting detection, robustness checks, and walk-forward and out-of-sample validation.
• Build the data foundations the agent reasons over, with provenance, validation and a consistent model of market entities, and make the system observable enough to diagnose non-deterministic failures in production.
• Shape what gets built by refining requirements with product, setting technical direction through RFCs and ADRs, and mentoring engineers moving into LLM systems.

What you bring

• 5+ years of backend or ML engineering, including at least one year building LLM-based systems in production. You understand agent failures one level below the framework and can fix them in your own code.
• A track record of eval-driven development: you can show a quality metric you kept honest over time and how you used it to improve an agent.
• Strong Python and production fundamentals (services, queues, streaming, observability, Kubernetes). You are comfortable following a bug into TypeScript or Go, and you treat pipelines, migrations and guards as real engineering.
• Application of statistical principles on backtesting methodology.
• A product mindset: you judge your work by what the user gets, and you take ambiguous goals to shipped, measured results.
• Working proficiency in English (B2+).

Nice to have

• Quant finance background: backtesting frameworks, market data and strategy research, including the pitfalls of backtesting indicator-based strategies.
• Experience with knowledge graphs, ontologies or semantic layers. RAG experience is a plus.
• Experience with Go.
• Software built on the receiving end of the MCP protocol, such as MCP servers or tools consumed by agents.
• High-load, low-latency systems in fintech or trading.
• Language model fine-tuning or training experience.

What we offer

• Ownership of services end to end, including on-call rotation for production support and incident response.
• An engineering culture where responsible use of AI-assisted development tools is expected and integrated into daily workflow.

About 3Commas

3Commas builds QuantPilot, an AI trading copilot bringing institutional-grade research tools to serious retail traders. The Applied AI team focuses on the agent engine that powers strategy research, generation and iteration.