Building the web data infrastructure that powers the next generation of AI applications.
Search Engineer
You'll own the retrieval pipeline that decides whether a query surfaces exactly the right document or buries it. In your first year, you'll redesign the core search architecture—moving beyond keyword matching into hybrid retrieval that blends sparse, dense, and learned ranking signals. This isn't tuning an off-the-shelf engine; it's engineering a system that crawls, indexes, and makes sense of the open web at a scale most search teams never touch. You'll ship improvements that directly change how AI models consume and reason over live web data.
What they're looking for
- 3+ years building production search or information retrieval systems, with deep fluency in at least one of Go, Rust, or Python
- Experience operating ElasticSearch at scale—index design, relevance tuning, and query performance optimization are second nature
- Comfortable reasoning about ranking end-to-end: tokenization, embeddings, hybrid retrieval, and the tradeoffs between precision and recall
- Strong systems intuition across message queues (Kafka, Redis), relational databases (PostgreSQL), and containerized infrastructure (Kubernetes, Docker)
- Bias toward shipping incrementally and measuring real outcomes—you'd rather run an A/B test than debate theory