The Opportunity :
We’re delighted to be working with a VC backed AI start-up currently in a period of growth that are that’s redefining how people learn - and they’re hiring.
Our client is a well-funded UK-based company operating at the intersection of large language models and learning science. Their mission is to deliver fast, reliable, and personalized educational experiences powered by cutting-edge AI. They’ve recently secured significant funding and are now looking for an AI/ML Engineer to join their founding team.
This role is ideal for someone who thrives on solving hard technical problems — particularly around reducing hallucinations in LLMs and optimizing latency for real-time learning applications.
What you’ll do:
* Develop and deploy techniques to minimize hallucinations in LLM-generated educational content
* Optimize latency across model inference, token streaming, and content delivery pipelines
* Build and evaluate custom datasets for grounding, factual accuracy, and curriculum alignment
* Collaborate with product, engineering, and learning science teams to ship production-ready AI features
* Stay current with research in LLM alignment, quantization, and retrieval-augmented generation
What we’re looking for:
* 4+ years experience in ML/AI engineering (LLMs, transformers, generative models)
* Strong Python skills and experience with PyTorch, Hugging Face, or similar frameworks
* Proven track record in latency optimization (e.g. batching, quantization, caching strategies)
* Familiarity with hallucination mitigation techniques (e.g. RAG, chain-of-thought, fact-checking)
* Bonus: experience in edtech, instructional design, or learning science