Job Details

Multiscale is building the world's first vertical-AI platform for microelectronics and materials science. Our MIND platform powers autonomous AI agents that design experiments, optimize manufacturing steps, and accelerate scientific discovery. You'll lead the charge in architecting and scaling these agentic systems.

What You'll Do

• Architect Multi-Agent Frameworks: Design and implement orchestration layers (e.g. LangGraph, AutoGen, OpenAI Swarm) so agents can plan, coordinate, and solve complex workflows.
• Develop MCP (Model Context Protocol): Build the server and APIs that enable agents to persist context, access tools, and share state across long-running tasks.
• Integrate Domain Logic: Partner with AI Scientists to embed materials-science and semiconductor-fab knowledge—enabling agents to call Python-based analysis routines and enforce safety guardrails.
• Prototype to Production: Rapidly build proof-of-concept agent chains; then harden them into scalable microservices (FastAPI) with monitoring (LangSmith) and container orchestration (Docker/Kubernetes).
• Drive Innovation: Stay on the cutting edge of autonomous AI—run experiments, benchmark new techniques (e.g. multi-agent RL), and influence our product roadmap.

What We're Looking For

• 7+ years software development in Python, with 3+ years on AI/ML systems.
• 2+ years building or shipping autonomous-agent solutions (LangChain, AutoGPT, CrewAI, etc.).
• Hands-on with LLMs (OpenAI, Claude, Llama) and prompt engineering best practices.
• Experience deploying microservices on AWS, GCP, or Azure using Docker & Kubernetes.
• Strong communicator: able to evangelize agentic design across engineering and science teams.

Nice to Have

• Open-source contributions or publications in multi-agent AI.
• Familiarity with semiconductor processes or materials-science workflows.
• GPU inference optimization experience (CUDA, Triton, vLLM).

• Mid-Senior level

• Full-time

• Information Technology

• Software Development