Lead Software Platform Engineer, MLOps (Remote)

TetraScience is hiring a Lead Software Platform Engineer to architect and scale AI/ML infrastructure for their Scientific Data and AI Cloud platform. This role owns the technical design of production-grade MLOps systems, including model and prompt lifecycle management, inference serving, security boundaries, evaluation, and observability. You will work closely with Applied AI, data engineering, and platform teams to build multi-tenant infrastructure for scientists at major pharmaceutical companies.

• Own the technical architecture of the AI/ML platform, serving as the design authority for multi-tenant infrastructure.
• Manage the end-to-end model and prompt lifecycle across Databricks MLflow and AWS Bedrock.
• Design the inference substrate for real-time and batch AI workloads, including routing, batching, and GPU capacity planning.
• Integrate LLMs into production with RAG, tool calling, MCP-based tooling, and agent runtimes.
• Embed security into the AI platform, including prompt-injection defenses, PII/PHI handling, and tenant isolation.
• Build evaluation and quality infrastructure, including offline/online eval harnesses, regression gates, and shadow deployment.
• Establish observability with SLI/SLO/SLA models and distributed tracing for probabilistic outputs.
• Contribute to infrastructure-as-code using CloudFormation and AWS CDK, and manage production readiness.