Design, build, and iterate AI and ML-powered ranking and relevance systems (including candidate generation, filtering, personalization, bandits, and learning-to-rank) that balance engagement, quality, and safety while meeting latency and reliability SLOs.
Develop features and embeddings for users, content, graph, temporal, and safety signals; establish feature contracts, offline backfills, and online freshness.
Stand up vector retrieval and RAG pipelines for semantic search, content understanding, and safety heuristics; tune retrieval quality, token/latency budgets, and evaluation.
Build and operate online inference and batch scoring services for LLMs and ML models; optimize throughput, cost, caching, and autoscaling; introduce GPU-aware scheduling where appropriate.
Partner with Data Platform on data movement and pipelines (batch and streaming) that power ranking models and analytics; ensure data quality, lineage, documentation, and observability.
Partner with Data Science to frame metrics and hypotheses, design trustworthy experiments, analyze results, and productionize improvements.
Productionize models end-to-end (training, evaluation, versioning, registries, rollouts, shadowing/canary, rollback) with strong CI/CD and infra-as-code practices.
Establish ML observability (data/feature drift, model performance, safety thresholds, online/offline eval, dashboards and alerting) to maintain quality over time.
Collaborate cross-functionally with product, design, platform/infra, safety, and CX to align on goals, roadmap, and risk management.
Contribute to engineering excellence through documentation, code reviews, reliability improvements, and mentorship to junior engineers.

Requirements

3+ years of ML engineering or recommender-systems experience (or equivalent), including shipping production models and improving them iteratively.
Strong software engineering fundamentals with proficiency in Python and SQL; experience building reliable services and data pipelines at scale.
LLM and AI inference experience, including prompt/program design, evaluation frameworks, agentic workflows, or online performance optimization.
Data pipelines and orchestration (batch and streaming), data quality/observability, and schema/feature contract management.
ML operations (model registries, feature stores, experiment frameworks, offline/online evaluation, rollouts, and monitoring).
Experimentation mindset with practical A/B testing design, metric selection, and rigorous analysis in partnership with Data Science.
Clear, pragmatic communication and the ability to collaborate across product, design, and engineering to drive outcomes.
A hunger to learn more about AI and ranking at production scale
BS/MS in Computer Science, Machine Learning, or equivalent practical experience.

Benefits

Full time offers from Coinbase include bonus eligibility + equity eligibility + benefits (including medical, dental, vision and 401(k)).

Applicant Tracking System Keywords

Tip: use these terms in your resume and cover letter to boost ATS matches.

Hard skills

machine learningrecommender systemsPythonSQLdata pipelinesML operationsA/B testingLLM inferencefeature engineeringCI/CD

Soft skills

communicationcollaborationmentorshipproblem-solvinganalytical thinkingexperimentation mindsetpragmatic approachcross-functional teamworkdocumentationengineering excellence

Certifications

BS in Computer ScienceMS in Machine Learning