Staff ML Infrastructure Engineer

About Playlab Playlab is a tech non-profit dedicated to helping educators and students become critical consumers and creators of AI. We believe that an open-source, community-driven approach is key to harnessing the potential of AI in education. We equip communities with AI tools and hands-on professional development that empowers educators & students to build custom AI apps for their unique context. Over 60,000 educators have published apps on Playlab – and the impact is growing every day. At Playlab, we believe that AI is a new design material - one that should be shaped by many to bring their ideas about learning to life. If you're passionate about building creative, equitable futures for students and teachers, we hope you’ll join us. The Role Playlab seeks a Staff Machine Learning Engineer to join our growing Engineering team. As a Staff ML Infrastructure Engineer, you'll be designing the systems that keep AI accessible as we grow - balancing cutting-edge capabilities with cost efficiency, powering research into what works in educational AI, and building toward a future where sophisticated AI can run anywhere in the world. Examples of the work • Build data pipelines that scrub PII, create research datasets, and power the research portal for educational AI studies • Architect the path toward self-hosted and on-device model deployments for privacy and global accessibility • Design and implement model orchestration systems that intelligently route requests across multiple AI providers (OpenAI, Anthropic, AWS Bedrock, open-source models) • Build cost optimization infrastructure - implement conversation compression, prompt caching, and smart model selection to keep AI accessible • Create comprehensive observability systems for ML operations - track costs, latency, quality, and usage patterns across thousands of applications • Design and implement infrastructure for fine-tuning and deploying custom models • Build monitoring and alerting systems that help us maintain reliability as AI interactions scale • And more… Expectations • Design, build, and maintain production ML infrastructure that balances performance, cost, and reliability • Own data quality and research dataset creation - ensure data is properly scrubbed, documented, and useful for research partners • Stay on top of ML infrastructure technologies and techniques - from model serving to cost optimization to observability tools • Work cross-functionally with ML engineers, backend engineers, and product to ensure infrastructure supports real needs • Balance innovation with operational excellence - experiment with new approaches while maintaining system reliability and data quality • Mentor engineers on ML operations, cost optimization, and production ML best practices Qualifications • 7+ years building production ML/data systems, with experience in ML operations and infrastructure • Strong experience with model serving, orchestration, and optimization in production environments • Proficient in Python and data pipeline technologies (Airflow, ETL tools, etc.) • Experience with cloud infrastructure (AWS preferred) and containerization (Kubernetes, Docker) • Experience with cost optimization strategies for LLM-based systems • Thrive in high-agency, high collaboration cultures • Great communication that makes working remote-first work Bonus Points For... • Experience in education or building in edtech • Experience with educational technology or mission-driven organizations • Experience with designing creative platforms • Experience with LiteLLM or similar model routing frameworks • Background in privacy-preserving ML or PII handling • Experience building research data infrastructure • Contributions to open source ML infrastructure projects Technologies Python, AWS, Kubernetes, Docker, Airflow, LiteLLM, PostgreSQL, Neo4J, Vector Databases, Terraform, Monitoring tools (New Relic, OpenTelemetry)