Research Intern - Systems for Reliable and Scalable AI Agents

Microsoft Microsoft · Big Tech · Redmond, WA +1 · Applied Sciences

Research Intern position focused on building reliable, safe, and efficient AI agents. The role involves exploring formal verification, reliability, safety, intelligent task handling, secure execution, and efficiency optimization for AI agents. It is a research-focused role within Microsoft Research, aiming to tackle complex challenges in AI agent systems.

What you'd actually do

  1. Formal Verification: Modeling and verifying AI agent's behavior to ensure correctness.
  2. Reliability & Safety: Ensuring agents execute tasks without failure or unintended side effects.
  3. Intelligent Task Handling: Designing agents that can efficiently manage complex workflows.
  4. Secure Execution: Guaranteeing agents operate in isolation and only perform authorized tasks.
  5. Efficiency Optimization: Minimizing the number of steps, step latency, and costly interactions with LLM backends.

Skills

Required

  • PhD program in Computer Science or a related STEM field

Nice to have

  • building scalable and reliable AI systems
  • develop original research agenda
  • collaborate effectively with other researchers and product development teams
  • Proficient interpersonal skills, cross-group, and cross-culture collaboration
  • think unconventionally to derive creative and innovative solutions

What the JD emphasized

  • AI agents
  • reliable, safe, and efficient AI agents
  • formal verification
  • reliability & safety
  • secure execution
  • efficiency optimization
  • tooling for developers

Other signals

  • AI agents
  • reliable, safe, and efficient AI agents
  • intelligent task handling
  • formal verification
  • reliability & safety
  • secure execution
  • efficiency optimization