Teams at Baselayer

Own production LLM-driven agent systems for business data enrichment, including classification, web discovery and verification, entity linking, legitimacy scoring, and risk signals. Build shared agent infrastructure, tools, evaluations, tracing, and cost controls. The role requires production Python engineering, browser automation, multi-provider LLM expertise, structured-output design, web scraping, and evaluation methodology. This is a hybrid role based in San Francisco with four in-office days weekly.

Build and maintain production ETL/ELT pipelines ingesting public records, web signals, and fraud telemetry. Develop data models and transformation layers using Dataflow, Spark, and Airflow; implement data quality, observability, and alerting; optimize warehouse performance, freshness, and cost; and collaborate with data science, ML, and product teams. Support security and regulatory standards for sensitive data, document systems, and communicate with technical and non-technical stakeholders.

Own end-to-end identity graph product features, including public-record ingestion, normalization, entity resolution, customer APIs, schema design, migrations, performance scaling, multi-tenancy, and access control. Build reliable systems for search, monitoring, webhooks, fraud detection, and KYC/KYB workflows while maintaining low latency under production load.