Senior AI Engineer, MapGPT

Accountabilities:: Define technical approaches for AI-powered products, establish measurement frameworks, and build evaluation systems for non-deterministic behavior. Develop datasets, hypotheses, evaluation criteria, and regression gates to determine whether system changes produce correct and reliable results. Continuously evaluate APIs, SDKs, data representations, and reference applications from the perspective of developers, agents, and end users. Identify issues such as incorrect parameter usage, integration anti-patterns, and other product gaps, recommend improvements, and drive fixes through to completion. Own MVP delivery against an agreed technical design while balancing long-term technical quality with the need to ship useful increments quickly. Build and maintain data pipelines and supporting tooling, including ingestion, conflation, entity resolution, quality checks, and batch and streaming processing. Evaluate external datasets, models, benchmarks, research, and open-source technologies, deciding when to adopt existing solutions versus building internally. Design feedback loops that turn product usage into data for continuous improvement and convert recurring failures into evaluation cases. Define the boundary between models and the tools they call, including what the model handles, what it delegates, and how context is maintained. Build and improve model harnesses and agent orchestration systems while addressing issues such as stale context, hallucinated arguments, partial success, and unbounded loops. Optimize systems against latency and cost targets through streaming, partial results, caching, model routing, and prompt design. Build internal developer tools and harnesses, including CLI and MCP-based tooling, and share reusable components across teams. Contribute to code and design reviews while mentoring engineers on evaluation practices and high-quality AI system development. Help investigate and solve emerging technical challenges where established approaches may not yet exist. Participate in an on-call rotation supporting system availability 24/7, including potential response requirements outside normal working hours and on weekends. Requirements: Bachelor's degree in a STEM discipline and 5+ years of software engineering experience with production ownership of services, pipelines, SDKs, or comparable systems. 2+ years of experience shipping LLM-backed features to real users in production environments with customers, error budgets, and on-call responsibilities. Strong data engineering foundation, including SQL, at least one distributed processing framework, and experience building pipelines where data accuracy and correctness are critical. Strong understanding of tool calling and agent orchestration, including common failure modes such as stale context, hallucinated arguments, silent partial success, and unbounded loops. Strong proficiency in Python or TypeScript, with the ability to read and work with code written in other languages when necessary. Direct experience or deep understanding of evaluation design for non-deterministic AI systems, including the ability to explain datasets created and the failures they identified. Working knowledge of more than one agent harness and an understanding of the strengths and limitations of different approaches. Experience diagnosing latency across distributed request paths. Strong judgment and comfort working with ambiguity, including the ability to deliver focused solutions while broader requirements are still evolving. Experience with geospatial data, including routing, geocoding, points of interest, address data, OpenStreetMap, or data conflation, is a plus. Public API or SDK design experience, particularly for developer-facing products, is a plus. Experience with MCP or similar tool transports is beneficial. Experience running evaluations in CI, whether through commercial tooling or internally developed systems, is a plus. Automotive, in-vehicle infotainment, CarPlay, or Android Auto experience is beneficial. Voice pipeline experience involving streaming ASR, TTS, barge-in, endpointing, or wake-word systems is a plus. Experience working under constrained compute, offline, or intermittent connectivity conditions is valuable. Experience taking a product through its first external integrations and addressing gaps identified through real-world customer usage is beneficial. Strong collaboration, communication, mentoring, and problem-solving skills. Benefits: Annual base compensation of $160,650–$217,350 for most U.S. locations, with compensation typically 5%–10% higher for U.S. locations with a higher cost of labor. Actual compensation determined based on qualifications, relevant skills and experience, role scope, potential impact, market conditions, and specific work location. Hybrid and remote work options, providing flexibility to work from home or periodically from an applicable office. Comprehensive private medical coverage for employees and dependents. Generous maternity and paternity leave. Fertility and family-building support. Lifestyle Spending Account contributions supporting health, wellness, and personal growth. Mental health support for employees and dependents. Flexible paid time away, company holidays, and generous absence policies. Dedicated paid volunteering time in addition to standard PTO. Retirement plans with competitive matching. Learning and development opportunities within a collaborative engineering environment. Opportunity to work on advanced AI systems spanning search, navigation, location intelligence, data, and agent-facing platforms. Exposure to complex production AI challenges involving LLMs, data pipelines, evaluation systems, and distributed architectures.