Machine Learning Engineer 2

Role overview

Build and operate systems that turn conversation data into reliable, searchable records of customer information for conversational AI. The work covers extraction, fast retrieval, and quality measurement, with production reliability and preventing stale or misattributed information as key concerns.

Responsibilities

- Build pipelines that extract structured customer memory from raw conversations, including tuning language-model prompts and data schemas. - Develop retrieval systems using embeddings, vector search, and ranking to return relevant information quickly. - Help create benchmarks and regression checks that measure memory quality before changes are released. - Investigate production issues across ingestion, extraction, and indexing using logs, metrics, and traces. - Partner with product and machine learning teams to turn customer needs into system changes and evaluation criteria.

Requirements

- Experience building backend services and the ability to learn unfamiliar languages, frameworks, and tools as needed. - Interest in understanding how services behave in production and operating them with rigor and accountability. - Experience with evaluation or quality measurement systems, such as regression detection or quality metrics, or willingness to develop that experience. - Working knowledge of observability, including structured logging and metrics for debugging distributed systems. - Ability to use AI-assisted development tools thoughtfully and critically.

Nice to have

Experience with Go or Python, event streaming, LLM-backed products, or applying machine learning in production.

Benefits and work setup

The role is remote, with occasional travel for team or project meetings. Hiring is unavailable in the San Francisco Bay Area locations specified in the listing. Benefits vary by location and may include healthcare, retirement savings, paid time off, parental leave, and wellness leave.