Drimva

Lila Sciences

Staff/Principal Engineer, Search & Retrieval

$224K–$336K

San Francisco, CA USA +1 · Principal · Full-time · Posted Oct 26, 2026

Location
San Francisco, CA USA
San Francisco, CA
Experience
8+ years of software development experience.
Education
BS, MS, or Ph.D. in Computer Science or a related field of study, or equivalent experience.

About the job

Own and build Lila Sciences’ shared search and retrieval platform for scientific information, from data ingestion through knowledge-graph and hybrid retrieval to cited answers and agentic research workflows.

What you'll do

  • Set the architecture and provide technical leadership for the shared retrieval platform across teams.
  • Build the scientific knowledge graph and combine graph, vector, and lexical retrieval into evidence-backed answers.
  • Own ingestion, retrieval quality, evaluation, and agentic search workflows end to end.

What they ask for

  • 8+ years of software development experience, including production ownership of a search, retrieval, RAG, or knowledge-graph system used by real users.
  • Production semantic-search experience with embedding models, vector indexes (HNSW/IVF), hybrid lexical and vector retrieval, and reranking.
  • Hands-on Python and modern backend framework experience; experience with cloud-native architectures including Kubernetes, Terraform, and AWS.
  • BS, MS, or Ph.D. in Computer Science or a related field of study, or equivalent experience.

Nice to have

  • Graph database experience; shipped RAG or semantic search in a regulated or enterprise context.
  • Experience with permission-aware retrieval, scientific document understanding, agentic retrieval, scientific data platforms, domain-specific corpora, or knowledge-graph construction and retrieval.

Pay

$224,000 — $336,000 USD (base salary)

Tools

Python, FastAPI, async Python, Kubernetes, Terraform, AWS, HNSW/IVF

AI focus

Build RAG, semantic and agentic search, a scientific knowledge graph, retrieval evaluation, and cited answers linking scientific literature to experimental records.

Model-written summary · the original posting is authoritative

First seen Oct 27, 2026 · Last verified Nov 5, 2026 (UTC)