Experience: 5–8+ years in backend, data infrastructure, or ML infrastructure, with clear evidence of shipping production systems rather than only experimental work.
- Strong ML platform / MLOps background: Hands-on experience building or operating model training, batch inference, real-time inference, release, and monitoring systems in production.
- Feature store depth: Experience with online/offline feature stores, point-in-time correctness, training-serving consistency, low-latency feature retrieval, and feature lifecycle/governance. Experience with Chalk or an equivalent framework is highly relevant.
- Cloud + infra fluency: Strong AWS-heavy experience, specifically with SageMaker, S3, Lambda, Kinesis, DynamoDB, Kubernetes, Docker, and Terraform.
- Programming strength: High proficiency in Python is required, and Go is a strong plus for this team's stack.
- Data platform fundamentals: Familiarity with Snowflake, Airflow, streaming systems, schema/data quality, and scalable ETL frameworks, as the MLP work sits closely with data-compute concerns.
- Reliability / observability mindset: Familiarity with SLOs, dependency graphs, drift detection, explainability, cohorting, and operational design.
- Cross-functional execution: Comfort partnering directly with Data Science, Risk, Product, Analytics, and Data Engineering teams.
To summarize, we are looking for contractors who can contribute directly to our ML Platform team's core systems—training pipelines, feature store, model deployment, and observability. Strong candidates should have real production experience with MLOps / ML infra, AWS, Kubernetes, Terraform, Python or Go, and ideally hands-on experience with feature stores, point-in-time correct data pipelines, and real-time or batch inference systems.
