We are looking for a pragmatic AI Engineer who can bridge solid software engineering practices with the fast‑evolving landscape of applied AI. You will help design, build, and evaluate AI systems that are reliable, measurable, and useful — not just demos. You will work closely with data scientists, software engineers, and product managers to ship AI solutions that make real impact. You believe in evaluation before hype, automation where it adds leverage, and simple systems that perform well in production in the life sciences industry. This is a hybrid working opportunity from either our Cambridge, UK or Leiden, Netherlands office.
Responsibilities
- Develop and deploy AI‑powered applications from prototype to production using modern LLM frameworks and APIs.
- Design evaluations and metrics that measure real‑world performance (accuracy, latency, UX quality, safety).
- Build and optimize Retrieval‑Augmented Generation (RAG) pipelines integrating vector databases and LLMs.
- Implement data pipelines, retraining workflows, and monitoring tools that keep models current and performant.
- Develop and implement context engineering strategies to improve output quality and control hallucination risks.
- Collaborate with product, platform and infrastructure teams to understand business needs and build systems that scale and fail gracefully.
- Apply agentic AI patterns (tool use, planning, reflection loops) pragmatically, choosing the right level of complexity for each use case.
- Champion good engineering practices: code reviews, testing, observability, CI/CD, reproducibility.
- Ensure AI solutions meet security, compliance and explainability standards.
Qualifications
- Passionate about building real AI? Skilled with LLMs, RAG pipelines, or agentic AI? Ready to deliver AI that actually works, not just demos? If yes, this role is for you!
- 3+ years of experience in data science or software engineering, ideally with 2+ years focused on AI engineering.
- Proficiency in Python and experience with LLM APIs (OpenAI, Anthropic, or similar services).
- Experience with GenAI frameworks and libraries (LangChain, LlamaIndex, Haystack, or Hugging Face Transformers).
- Experience with vector databases such as Pinecone, FAISS, Weaviate and RAG pipeline implementation.
- Proven ability to evaluate AI systems through custom evaluations, benchmarking or data curation.
- Developed and deployed AI‑powered applications end‑to‑end, overseeing all phases from initial concept, prototyping & design through implementation, testing, and production launch.
- Familiarity with agentic AI concepts (tool‑use, planning, multi‑step reasoning), with a practical sense for when simpler systems suffice.
- Strong grounding in software engineering principles (version control, modular design, testing) and commitment to measurement‑driven iteration.
- Excellent communication skills and the ability to explain complex AI solutions to non‑technical stakeholders.
- Strong understanding of AI ethics, bias mitigation, and responsible AI practices.
Nice to Have
- Experience with model fine‑tuning, knowledge graphs, or multi‑modal AI systems.
- Familiarity with AWS services for scalable GenAI deployment and MLOps practices.
- Demonstrated experience in direct client communication.
About Our Company
Work shouldn't just be something we do; it should have a purpose. At Envision, we believe in creating life‑changing outcomes through the work we do with our clients, giving back to our communities, while creating a company culture where our people thrive. We believe success starts with a workplace where everyone feels valued, supported and empowered to grow.
Vision & Mission
Our Vision: To unleash the power of combined intelligence to accelerate patient access to life‑changing treatments. Our Mission: Delivering smarter and faster solutions to create, communicate and commercialize value for our clients.