Python Insfrastructure Engineer - Model Evaluation

* Design, build, and optimize high-performance Python systems supporting AI data pipelines and model evaluation workflows * Develop full-stack tooling and backend services for large-scale data annotation, validation, and quality control * Build and maintain evaluation harnesses that integrate with ML inference frameworks * Improve reliability, performance, and safety across existing Python codebases * Instrument systems with observability and metrics collection to monitor reliability and model performance * Identify bottlenecks and edge cases in data and system behavior, and implement scalable fixes * Collaborate with data, research, and engineering teams to support model training and evaluation workflows * Participate in synchronous design reviews to iterate on architecture and implementation decisions

Who You Are

* Native or fluent English speaker with clear written and verbal communication skills * Full-stack developer with a strong systems programming background * 3–5+ years of professional experience writing production-grade Python * Experienced building evaluation harnesses for ML models and integrating with inference frameworks * Solid background in observability, metrics collection, and monitoring for production systems * Self-motivated and reliable — able to commit 20–40 hours per week

Nice to Have

* Prior experience with data annotation, data quality, or evaluation systems * Familiarity with AI/ML workflows, model training, or benchmarking pipelines * Experience with distributed systems or developer tooling * Background in MLOps or AI infrastructure

Why Join Us

* Work directly on cutting-edge AI projects alongside leading research labs * Fully remote and flexible — structure your work week around your life * Freelance autonomy with the depth and consistency of meaningful, long-term technical work * Make a tangible impact on how next-generation AI models are evaluated and improved * Potential for ongoing work and contract extension as new projects launch