Hired into a specific Mapbox team but expected to work across teams on Search, Places, Location/Navigation Intelligence, and Platform problems. Own technical design and delivery of multi-component AI systems: define correctness and evaluation frameworks for non-deterministic behavior, build and maintain data pipelines (ingestion, conflation, entity resolution, batch/streaming jobs), instrument systems for reproducible failures, design model vs tool boundaries and model harnesses, and meet latency/cost targets. Requires 5+ years software engineering experience and 2+ years shipping LLM-backed features; strong data engineering skills (SQL, distributed processing), fluency with tool calling/agents, Python or TypeScript, and participation in on-call rotations.

Senior-level software engineer responsible for owning technical design and delivery of multi-component AI systems across Mapbox teams. Responsibilities include building evaluation frameworks for non-deterministic systems, shipping LLM-backed features, developing data pipelines (ingestion, conflation, entity resolution, batch/streaming), instrumenting systems for reliability, and participating in on-call rotations. Geospatial and voice pipeline experience, API/SDK design, and MCP familiarity are listed as nice-to-have.

Lead engineering work on highly trafficked recommendation systems for the routing cost engine, including designing/building ML and data pipelines, leveraging geospatial algorithms, partnering with data scientists, building automated QA and data exploration tools, mentoring engineers, and participating in an on-call rotation. Requires ~8+ years backend and ML experience and familiarity with Python, distributed processing (Hadoop, Spark, Airflow, Dask), cloud infrastructure (e.g., AWS), and reinforcement learning.