Lead Software Platform Engineer, MLOps

Role

Lead Software Platform Engineer (MLOps) working at the intersection of distributed systems and MLOps. You will own and scale TetraScience’s AI and data infrastructure as a multi-tenant product that customers build their science on.

Responsibilities

• Own the technical architecture of the AI/ML platform, including the customer-facing service/API surface and internal platform interfaces.
• Own the end-to-end model and prompt lifecycle across Databricks MLflow and AWS Bedrock (registration, versioning, asset bundles, staged promotion, rollback, and multi-model serving).
• Design the inference substrate for both real-time and batch workloads (routing, batching, caching, concurrency control, GPU/accelerator capacity planning, large binary input handling like instrument images, and graceful degradation under load).
• Integrate models and LLMs into production using RAG, and architect the agentic layer (tool/function calling, MCP-based tooling, and agent runtimes), deciding what belongs in the platform vs. downstream applications.
• Build security into the platform: guardrails, defenses against prompt injection and tool abuse, handling PII/PHI, and enforcing hard tenant data boundaries across prompts, retrieval, and tool calls.
• Build evaluation and quality infrastructure to make AI shippable, including offline and online eval and related observability/quality mechanisms.

Requirements / Qualifications

• Experience shipping AI/ML infrastructure as a multi-tenant product (not internal tooling).
• Demonstrated ability to translate ambitious scalability and cost targets into technical strategy.
• Ability to operate in a regulated environment with customer compliance obligations.

Nice-to-haves

• Experience with production LLM/agent systems consumed by highly regulated scientific/health/pharma users.
• Deep familiarity with security, tenant isolation, and robust evaluation for AI workflows.