Product Engineer — Scrape

PRODUCT ENGINEER — SCRAPE You'll own Firecrawl's flagship product — the scrape endpoint that turns any URL into clean, LLM-ready data with a single API call. It's the product 100k+ developers know us for, the one that put Firecrawl on the map, and the one every new AI app reaches for when it needs the web as input. Your job is to make it unbeatable: faster, more reliable, better formatted, and more delightful to integrate than anything else on the market. This isn't a maintenance role. Scrape is the front door — and the front door has to be the best part of the house. At a 26-person company, the gap between "the scraper works" and "developers can't imagine using anything else" is exactly one person. You're that person. Salary Range: $180,000 to $290,000/year (Range shown is for U.S.-based employees in San Francisco, CA. Compensation outside the U.S. is adjusted fairly based on your country's cost of living.) 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 shipping developer-facing products — ideally in scraping, crawling, browser automation, or data infrastructure Visa: US Citizenship/Visa required for SF; N/A for Remote ABOUT FIRECRAWL 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 120k+ 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. WHAT YOU'LL DO Own the scrape product end-to-end. Scrape is the endpoint developers integrate first and the one they depend on most. You own how it feels — response format, latency, reliability, error messages, the markdown quality, structured extraction, every parameter, every edge case. If a developer has a bad time with /v1/scrape, that's on you. If they have a great one, that's on you too. Make "just works" actually true. The web is messy. JavaScript-heavy SPAs, anti-bot walls, dynamic content, infinite scrolls, weird charsets, broken HTML. Developers don't want to know about any of it — they want clean markdown. Your job is to push the "just works" rate from great to unbeatable, one long-tail failure mode at a time. Obsess over the output, not just the fetch. Pulling the HTML is table stakes. The product is the markdown — its structure, cleanliness, fidelity to the source, and how well it drops into an LLM prompt. You'll make calls about what gets stripped, what gets preserved, how tables and code blocks are handled, when to flatten vs. when to keep structure. These are product decisions disguised as engineering ones. Ship structured extraction that developers trust. Schema-based extraction, JSON mode, prompt-based extraction — developers use these to skip the LLM call entirely. They have to be reliable enough to build on. You'll iterate on the extraction surface until developers stop writing their own post-processing. Dogfood relentlessly. You build with the API before you ship changes to it. You feel the friction first. You read every GitHub issue, every Discord thread, every support ticket that touches scrape — not because someone asked you to, but because that's where the product signal lives. Run fast product experiments. Form a hypothesis about what would make scrape better, instrument it, ship it, measure it, decide quickly. You're comfortable making calls with imperfect data because waiting for perfect data means shipping nothing — and the competition isn't waiting. Raise the bar on developer experience. Firecrawl's users are technical. They notice when response formats drift, when error codes are unhelpful, when docs lag behind behavior. You notice too — and you fix it before they have to ask. WHAT WE'RE LOOKING FOR Obsessive about developer experience. You think about DX the way a designer thinks about pixels. Latency, response structure, error messages, API ergonomics, markdown quality — 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. Deep instincts for scraping and the messy web. This isn't abstract to you. You know why headless browsers fail, what anti-bot systems actually do, when to render JS and when not to, why some sites need stealth and others don't. You've felt the pain of a flaky scraper at 3am and you have opinions about how to build something that doesn't break. Speaks both product and engineering fluently. You can read a rendering pipeline and understand its implications for the developer experience. You can write the API spec and implement it yourself. You don't need a PM to tell you what matters. You connect the dots from "this 99th percentile latency