AI Context & Data Infrastructure Engineer — On-Site SF

Devopsroles; San Francisco (CA); On-site; USD 180,000 - 260,000; Full time

Job summary

Town is building the most personalized AI assistant for everyone, with a focus on retrieval-based context. You will architect the shared search layer that powers the Townie’s context, enabling fast access to memory, documents, and relationships across the product.

This greenfield work demands a strong systems thinker who can deliver scalable, cost-effective solutions from prototype to production, in a fully in-person environment at the SF office.

Qualifications

• Significant hands-on experience with lexical and semantic search, BM25, embeddings, and vector indexes.
• Experience designing scalable search/data infrastructure for production systems.
• Ability to reason about latency, cost, and freshness in large-scale retrieval.

Responsibilities

• Build the search and retrieval layer powering Townie's context at the moment of need.
• Combine lexical and semantic search, balancing vector vs lexical approaches and ranking.
• Design a durable data model to support fast, relevant context retrieval.
• Develop real-time streaming and batch pipelines to keep data fresh and correct.
• Implement indexing and storage layers for fast, scalable access.
• Lay groundwork for a knowledge graph connecting people, companies, projects, and relations.

Skills

Lexical search

Semantic search

Vector search

Data infrastructure

Distributed systems

Tools

BM25

Embeddings

HNSW

Python

Job description

Town is building the most personalized AI assistant for everyone, with a focus on retrieval-based context. You will architect the shared search layer that powers the Townie’s context, enabling fast access to memory, documents, and relationships across the product.

This greenfield work demands a strong systems thinker who can deliver scalable, cost-effective solutions from prototype to production, in a fully in-person environment at the SF office.