Nonprofits do some of the most important work in the world, and most of them are still managing grants in spreadsheets. We’re fixing that.
Instrumentl is a profitable, hypergrowth, YC-backed SaaS platform building the operating system for grant-funded organizations. More than 5,500 nonprofits use Instrumentl to discover, track, and win grant funding, from local community organizations to the San Diego Zoo and the University of Alaska. Collectively they’ve moved over $1 billion through our platform.
We’re doubling year over year, customers love us (NPS 65+, Ellis PMF 60+), and we’re hiring people who want to build something that matters.
About the role:
What you will do
- Build tool-using LLM systems that plan, call tools, and run multi-step workflows for tasks like grant discovery, data ingestion, and research assistance.
- Build agentic data-processing pipelines that crawl the web, pull and dedupe messy source data at scale, and structure it into clean, queryable databases other teams build on.
- Turn prototypes into resilient production services with clear fallback, cost, and latency budgets.
- Own RAG end to end: ingestion, chunking and embedding strategy, hybrid retrieval, re-ranking, citations, and grounding.
- Build ranking and scoring systems that match grants to nonprofits, universities, and foundations using complex relevance techniques.
- Continuously improve recall and precision and keep indices healthy as the dataset grows.
- Stand up evaluation and observability so our AI is grounded, safe, and cost-effective, and treat LLM behavior as non-deterministic by design rather than as a regular API.
- Partner directly with founders and your pod on undefined, complex problems with real autonomy.
- Write clear, maintainable, well-tested code and build reusably so your work expands across teams.
Build agentic AI systems and ship them to production
Own RAG and ranking end to end
Ship safely and raise the bar
What we're looking for:
- 7+ years of professional software engineering experience, with deep, recent, multi-year Python and strong relational database and schema design skills.
- Solid CS fundamentals and a demonstrated track record of owning complex systems end to end, from design through production reliability.
- At least 1 year of hands-on experience building with modern LLMs (as an IC).
- Real RAG depth: hybrid search (keyword plus vector), re-ranking or fusion methods, and grounded citations, tuned in production rather than read about.
- Hands-on with at least one of LangChain, LangGraph, or LlamaIndex.
- Vector databases beyond pgvector (Pinecone, Qdrant, Milvus).
- Built end-to-end agentic data-processing systems (crawling, dedup, structuring) with whole-system ownership.
- Evaluation and observability for AI systems: golden datasets, precision vs. recall, LLM-as-judge, and drift monitoring.
- Ruby on Rails (our core platform is on Rails), deep SQL, and experience with AWS or GCP, Docker, and CI/CD.
- Startup experience and comfort operating in fast, scrappy, low-process environments.
Required
Compensation & Benefits
- 100% covered health, dental, and vision insurance for employees (50% for dependents)
- Generous PTO, including parental leave
- 401(k)
- Company laptop and home-office stipend
- Bi-annual company retreats
- Instrumentl is evolving rapidly. You’ll always have new challenges and opportunities to grow here.
For US-based candidates, the target salary range for this role is 175,000 - $220,000 USD, plus equity. Final compensation is determined based on experience, skillset, scope of responsibility, interview performance, and geographic location. We’re committed to paying competitively and equitably.
For candidates based in Canada, compensation varies by province and will be shared by your recruiter early in the process.
Benefits



