You'll own the developer-facing search experience at Firecrawl — taking the retrieval and ranking improvements coming out of research and shipping them as a product developers can't stop using. This isn't a pure research role and it isn't a pure backend role. You sit at the intersection: you understand how the systems work deeply enough to improve them, and you care about how they feel to use obsessively enough to make them great. At a 26-person company, the gap between research and shipped product is exactly one person. You're that person.
Salary Range: $180,000–$290,000/year (Range shown is for U.S.-based employees. Compensation outside the U.S. is adjusted fairly based on your country's cost of living. You can explore how we calculate this here: https://www.firecrawl.dev/careers/compensation.)
Equity Range: Up to 0.15%
Location: San Francisco, CA or Remote (Americas, UTC-3 to UTC-10)
Job Type: Full-Time
Experience: 3+ years in applied RL, ML engineering, or model training — with production systems
Visa: US Citizenship/Visa required for SF; N/A for Remote
Firecrawl is the easiest way to extract data from the web. Developers use us to reliably convert URLs into LLM-ready markdown or structured data with a single API call. In just a year, we've hit 8 figures in ARR and 100k+ GitHub stars by building the fastest way for developers to get LLM-ready data.
We're a small, fast-moving, technical team building essential infrastructure superintelligence will use to gather data on the web. We ship fast and deep.
Ship search improvements that developers notice. Take retrieval and ranking improvements from research and turn them into product changes that make developers say "this just works." You know that a 200ms latency improvement isn't just a benchmark win — it's a better product. You ship the whole thing: the API change, the docs update, the example that makes it obvious.
Own the search API end-to-end. You're responsible for how Firecrawl's search endpoint feels to integrate, use, and build on. That means response format, latency, error handling, pagination, filtering — every surface a developer touches. You're the person who notices when something is confusing before a user files a GitHub issue about it.
Dogfood relentlessly. You build things with the API before you ship them. You feel the friction before your users do. You read every GitHub issue, every Discord thread, every support ticket that touches search — not because someone asked you to, but because that's where the product signal lives.
Translate research into product decisions. You work closely with the Search/IR and RL Research Engineers. You understand their work well enough to make good product calls about what to prioritize, what to expose to users, and what to keep under the hood. You ask good questions. You push back when something technically elegant would make the API worse.
Run fast product experiments. You form a hypothesis about what would make search better for developers, instrument it, ship it, measure it, and decide quickly. You're comfortable making calls with imperfect data because waiting for perfect data means shipping nothing.
Raise the bar on developer experience. Firecrawl's users are technical. They have high standards. They notice when response formats are inconsistent, when error messages are unhelpful, when documentation doesn't match behavior. You notice too — and you fix it before they have to ask.
Obsessive about developer experience. You think about DX the way a designer thinks about pixels. Latency, response structure, error messages, API ergonomics — these things matter to you on a visceral level. You've built APIs that developers loved and you know the difference between an API that works and one that delights.
Speaks both product and engineering fluently. You can read a ranking algorithm and understand its implications for the search experience. You can write the API spec and implement it yourself. You don't need a PM to tell you what matters or an ML engineer to explain why a retrieval change is significant. You connect those dots on your own.
Hands-on builder who ships. You write code. You own features from design to deployment. You're comfortable with ambiguity and you don't need a perfectly scoped ticket to make progress. You ship something, learn from it, and iterate.
Has a feel for search as a product. You've thought seriously about what makes search good — not just fast or accurate, but genuinely useful. You understand the difference between recall and precision and why developers care about both. You have intuitions about query understanding, result ranking, and when semantic search beats keyword search — and you've built products that put those intuitions to work.
Brings production instincts. You've operated systems under real load. You know what breaks first, how to instrument what matters, and how to make good latency/quality tradeoffs. You're not just building features — you're building infrastructure developers depend on.
Backgrounds that tend to do well: Engineers who've owned search or discovery features at developer-tools companies. Full-stack engineers with a strong backend bias who've shipped APIs used by thousands of developers. Engineers from search infrastructure teams who got frustrated by the distance between their work and the user experience. People who've built on top of Elasticsearch, Vespa, or vector databases — and cared enough about the product layer to go deeper than the query interface.
Great engineers who don't care about DX. If you build technically excellent systems but think API ergonomics and documentation are someone else's problem, this isn't the role. The product experience is part of the job — not an afterthought.
People who need a PM. There's no product manager between you and the work. You define what good looks like, you decide what to prioritize, and you own the outcome. If that's uncomfortable, you'll struggle here.
Specialists who only work on one layer. If you're only interested in backend systems and tune out when the conversation shifts to how something is exposed to developers — or vice versa — this won't be a fit. This role requires you to hold both.
Slow shippers. The research team will produce improvements faster than a slow product cycle can absorb them. We need someone who can take something from "this ranking model is better" to "this is live in the API with docs and an example" in days, not sprints.
People who don't use the product. If you're not the kind of engineer who builds side projects with APIs like ours, reads the docs critically, and notices when something feels off — you'll miss the signal that makes this role work.
We operate at an absurd level of urgency because the window for what we're building won't stay open forever. If that excites you, keep reading. If it doesn't, no hard feelings — but this role probably isn't for you.
Salary that makes sense — $180,000–$290,000/year, based on impact, not tenure
Own a piece — Up to 0.15% equity in what you're helping build
Generous PTO — 15 days mandatory, anything after 24 days, just ask (holidays excluded); take the time you need to recharge
Parental leave — 12 weeks fully paid, for moms and dads
Wellness stipend — $100/month for the gym, therapy, massages, or whatever keeps you human
Learning & Development — Expense up to $1,000/year toward anything that helps you grow professionally
Team offsites — A change of scenery, minus the trust falls
Sabbatical — 3 paid months off after 4 years, do something fun and new<
Firecrawl