• Design, develop, and optimize scalable Machine Learning models for advertising intelligence and forecasting systems • Analyze large-scale historical and real-time datasets to improve forecasting accuracy and monetization strategies • Build and maintain distributed data processing pipelines using Spark and related Big Data technologies • Develop production-ready ML solutions using Python within AWS cloud infrastructure • Research, evaluate, and implement Machine Learning algorithms suitable for high-load AdTech environments • Improve model performance, scalability, reliability, and operational efficiency • Collaborate with Data Engineers, Product Managers, and distributed engineering teams to deliver end-to-end ML solutions • Contribute to ML architecture decisions, experimentation approaches, and engineering best practices • Monitor, validate, and optimize ML model quality and forecasting performance in production environments • Produce technical documentation related to ML pipelines, models, and distributed systems • Mentor engineers and support technical growth within the team • Drive innovation in Machine Learning and AdTech technologies

• Own the research-to-production pipeline • Partner directly with research scientists to productionize new models • Build and extend the tooling and abstractions • Design and own model release gates • Optimize models and serving for production • Strengthen the build and delivery layer for models • Establish benchmarking and validation • Build the feedback loop

• Build and iterate on consumer-facing AI features powered by large language models (LLMs) and related generative AI systems • Collaborate with engineers across the AI engineering stack — prompt engineering, agentic workflow optimization, evaluation and more • Run structured experiments (e.g., A/B tests, offline evaluations) to measure impact on engagement, quality, and trust metrics • Partner closely with product and engineering to translate user needs into scalable AI-powered solutions • Strengthen model reliability through monitoring, guardrails, quality analysis, and failure mode investigation • Optimize the latency, cost, and scalability of production AI systems

Role Description Our client, a Silicon Valley-based industry leader in retail AI solutions, is now looking for an experienced MLOps Engineer to join its expanding team. They are growing their ML Platform function. The successful candidate will be an early member with broad ownership across their deep learning inference infrastructure — GPU serving, pipeline performance, observability, and safe model rollout. High autonomy: the right person will shape both the systems and the practices as they scale. Responsibilities: - Design, build, and operate large-scale deep learning inference pipelines processing millions of images per day across multiple model services - optimized for low latency, high throughput, GPU efficiency, and compute cost. - Plan and manage GPU fleet capacity - drive GPU utilization and cost efficiency (reservations, right-sizing, blue/green headroom). - Own observability for inference services: metrics, tracing, dashboards, and alerting on latency, throughput, accuracy, and availability. - Safely roll out deep learning models via staged/canary releases that protect accuracy and latency, with fast and reliable rollback. - Detect and respond to latency regressions, accuracy degradation, and unavailability - lead incident response and root-cause analysis; uphold a 99.9% uptime SLO (on-call rotation). - Partner with engineering teams on backend data needs - ensure data persists in usable formats for front-end, middleware, diagnostics, and model training/eval. - Build tooling for discoverability and access to the company's datasets across geographies and data formats. Qualifications - 5+ years building and operating production systems spanning ML serving, backend services, and infrastructure. - Bachelor's Degree or higher in CS, EE - or equivalent practical experience. - Excellent programming skills in Python. - Strong experience with containerization and orchestration: Docker, Kubernetes, Helm. - Great understanding of SQL, KeyValue stores, networking, distributed systems, operating systems, data structures, algorithms, and software engineering practices. - Experience with streaming / event-driven pipelines using Kafka (or equivalent). - Experience with high-availability operations management, including deployment automation and rollback strategies. - Startup mentality, team player and willing to work 40+ hours a week. - Advanced English skills (written and spoken). Requirements - Nice to have: Ray / Ray Serve (or comparable distributed compute/serving frameworks). - MongoDB / Atlas at scale, including vector search (or similar NoSQL/KV stores). - Observability stack: Grafana / Prometheus. - GPU cost optimization and capacity planning. Benefits - Competitive salary. - Stock options package. - Quarterly team retreats.