Founding Engineer
Help us read the entire AI answer layer as structured data. You'll build the scraping and parsing pipelines, the public API and the infrastructure that keeps them fast and honest as we scale — with the ownership only an early engineer gets.
# what you'd own from day one { "role": "Founding Engineer", "focus": ["backend", "full-stack"], "you_build": [ "scraping + parsing pipelines", "the public API", "infra that scales" ], "team_size": "small & growing", "ownership": "high", "stage": "early" }
Turn a moving target into clean, reliable data.
cloro is an API that turns AI answer engines and search results into structured data. ChatGPT, Gemini, Perplexity, Copilot, Grok, Google AI Overview, AI Mode and classic Google Search all render differently, rate-limit hard and change constantly. Our job is to absorb that chaos and hand developers one clean, stable schema.
As a founding engineer you sit at the center of that: fetching live results without getting blocked, parsing messy HTML and answers into typed records, and serving it all through an API that stays fast under real volume. You'll make architectural calls that outlive this quarter, and you'll see them run in production days — not months — later.
Where this role sits
Close to the metal and close to the product. The pipeline you write feeds every customer request; the API you shape is the product's entire surface.
Build the engine, then keep it honest at scale
Build the fetch layer
Design resilient scraping across AI answer engines and search — rendering, rotation, retries and freshness — so a query reliably comes back as data, not a block page.
Parse into one schema
Extract answers, citations, ranked links and named brands from wildly different sources and normalize them into a single typed shape customers can trust.
Ship the API
Own endpoints, request validation, error semantics, usage metering and docs — the developer experience that makes cloro feel like a database, not a scraper.
Scale the infrastructure
Move from "works on my batch" to thousands of concurrent queries with predictable latency — queues, workers, storage and the observability to see it all.
Set the foundations
Choose the stack, wire up CI, testing and deploys, and establish the patterns the next engineers will build on. Early decisions, lasting impact.
Talk to users
Sit close to the developers using cloro, turn their pain into roadmap, and ship fixes fast. Product sense here is part of the engineering job.
# you'll thrive here if… match candidate: ships_end_to_end = true loves_backend = true comfortable_with_ambiguity = true cares_about_data_quality = true # nice, not required scraping_at_scale = "a plus" distributed_systems = "a plus"
Range, ownership, and a bias to ship
Titles and years matter less than what you can build and how you think. If most of this sounds like you, we'd love to talk.
- Strong backend engineering with real full-stack range — you can carry a feature from data model to endpoint to a working UI
- You've built and operated systems that talk to the messy outside world: HTTP, HTML, flaky third parties, rate limits
- Comfort with data-heavy work — parsing, normalization, queues, databases and caching
- You care about correctness and observability, not just "it returned 200"
- You're at home in an early, ambiguous environment and you turn vague problems into shipped software
- Bonus: scraping / anti-bot experience, distributed systems, or having run an API product before
Early enough to matter, real enough to run
Search is being rewritten in front of us, and the teams keeping score need a clean way to read it. That's the problem we're building infrastructure for — and it's just getting started.
A problem with a tailwind
The answer layer is new, changing weekly and hard to structure. That's exactly the kind of durable problem worth building deep infrastructure around.
Ownership, not tickets
You'll own whole surfaces, make the calls, and watch them ship. Small team, short path from idea to production, credit where it's due.
Engineering as the craft
Pragmatic, curious, low-ego. We prototype fast, keep what works, write it down, and treat data quality as a first-class feature.
Want to build cloro with us?
Email us a few honest lines about what you've built, links to code or projects you're proud of, and why this problem interests you. No form, no cover-letter theatre — just reach out.