Senior Backend Engineer, Instrumentl

• 7+ years building and shipping production backend systems in Python (FastAPI, Celery, or equivalent), taking features from prototype to production with real reliability practices like tests, observability, and rollback
• Hands-on experience building LLM features in production: tool and function calling, multi-step agent workflows, and the guardrails and evals that keep them grounded, safe, and cost-effective. This is the core of the role
• Strong data fundamentals: SQL, schema design, and building pipelines that power retrieval and evaluation
• Thrives in a fast, scrappy startup environment with high ownership and a bias for action, speed, quality, and simplicity

Desirable

• We're hiring a Senior Backend Engineer to own AI features end to end, from rapid prototype to production and the evaluation that keeps them honest
• You'll build the APIs, tool-using agents, and RAG pipelines that turn frontier LLMs into grant discovery, application drafting, and research tools our 5,500+ nonprofits rely on every day
• It's a high-ownership seat on a small team, where what you ship reaches customers fast and you help shape how we build AI here
• Ship AI to production:
• Build tool-using LLM agents (task planning, function and tool calling, multi-step workflows, guardrails) for grant discovery, application drafting, and research assistance
• Turn prototypes into resilient, observable services with clear SLAs, rollback and fallback strategies, and cost and latency budgets
• Stand up evaluation and observability so our AI stays grounded, safe, and cost-effective
• Build trustworthy backends:
• Write high-quality, thoroughly tested code across the backend and the data pipelines that power retrieval and evaluation
• Contribute to reliability practices: alerts, dashboards, and incident response
• Collaborate and raise the bar:
• Partner with Product, Design, and GTM on scoping, UX, and measurement
• Run experiments (A/B, canaries), interpret results, and iterate
• Raise engineering standards through clear, maintainable code, tests, docs, and thoughtful review