Intern: Agentic AI Engineering & Research Assistant

The opportunity

We’re building a dedicated agentic AI platform that enables fully autonomous research assistants. These agents will interpret user queries, break them into tasks, select the right tools, collaborate with one another and the user, and deliver structured, actionable results — all without requiring traditional dashboards or manual model-building.

You’ll work closely with Joel Strickland (Head of Agentic AI) on one of the most exciting problems in AI: making agents think like scientists.

This role blends engineering and research: designing pipelines, experimenting with multi-agent behavior, integrating memory and planning systems, and helping build the foundation of a transformative product in applied AI.

Areas of research

Agent Research & Prototyping: Explore and implement architectures for agent collaboration, planning, and decision-making

Tooling Infrastructure: Build wrappers around APIs (esp. Alchemite™) so agents can use modeling, analysis, and optimization tools autonomously

Multi-Agent Workflows: Help define and orchestrate interactions between agents with different roles (planner, modeler, analyst, communicator)

Memory & Planning: Integrate document retrieval, long-term memory, and planning modules to support iterative reasoning

Interface Prototyping: Contribute to a lightweight, text-based prompt interface that showcases end-to-end task automation

User-Centered AI: Ensure that agents deliver insight to the user — not the other way around — by proactively surfacing results, next steps, and explanations

What We’re Looking For

You’re an ambitious, systems-thinking engineer or researcher with a deep curiosity about LLMs, multi-agent systems, and the future of AI-assisted discovery.

Essential:

• Strong Python skills and experience with APIs or orchestration frameworks
• Interest in LLM-based agents, autonomous workflows, or AI planning
• Willingness to dive into research papers, experiments, and fast iterations
• Clear communicator, comfortable working independently and collaboratively
  

Bonus Points For:

• Experience with LangChain, Vertex AI, Smolagents, HuggingFace, MCP servers, or similar
• Familiarity with RAG pipelines, embedding models, or memory systems
• Knowledge of scientific modeling, optimization, or data analysis workflows
• Past work or coursework in AI research, agent design, or cognitive architectures
• Background in modeling, optimization, or scientific computing
  

What You’ll Gain

• Deep experience at the frontier of applied AI and autonomous systems
• Mentorship from AI leaders shaping the next generation of R&D tools
• Real-world engineering + research exposure in agentic infrastructure
• Opportunity to extend the role or join full-time based on contribution
• The chance to build something truly new — agents that think, reason, and solve real-world scientific problems
  

How to Apply

Send us a short note on why this project excites you, along with your CV and (optionally) links to GitHub, research, or relevant projects.

🚀 Start Date: ASAP
📅 Applications reviewed on a rolling basis

About Intellegens

At Intellegens we enable our customers to apply advanced machine learning methods, accelerating innovation in materials, chemicals, manufacturing, and beyond. Our software can extract hidden value from real-world experimental and process data, serving organisations such as NASA, Johnson Matthey, Boeing, and AstraZeneca. The results are: faster R&D; greener, more cost-effective, better products; and more efficient use of resources. Originally a spin-out from the University of Cambridge, we are at an exciting phase in our development, with well-established products and customers, ambitious growth plans including a recent move to our new headquarters in Cambridge, whilst expanding our team into the US. We push the boundaries of machine learning, while working on interesting projects in a friendly, informal, and motivational work environment.