Brainpool is a fast growing AI start-up, providing custom AI services for business since 2017. Brainpool network of 500 top-level AI and Machine Learning experts supporting delivery of our projects include PhD and MSc-level scientists from top universities such as UCL, Oxford, Cambridge and Harvard. Brainpool provides companies with end-to-end AI solutions, starting from strategy development, planning all the way to implementation.
This role will involve working alongside the CTO and a team of engineers to develop cutting-edge AI solutions for client projects and internal company products. You'll contribute to key deliverables such as AI scoping programmes, PoC development, and production-ready AI systems, with a significant focus on Brainpool's LLM-powered applications like the AI Quote Generation system detailed in our recent client proposal.
Key Responsibilities
- Design and implement robust RAG (Retrieval-Augmented Generation) pipelines for processing complex document collections and technical documentation
- Develop LLM integration systems with AWS Bedrock for models like Claude and Mistral
- Create evaluation frameworks for benchmarking and improving LLM performance
- Build and optimize FastAPI-based REST endpoints with Swagger documentation for AI services
- Implement vector database solutions using technologies like Qdrant
- Contribute to containerized development and deployment environments using Docker and AWS ECS
- Participate in prompt engineering and enhancement of HyDE-based retrieval systems
- Collaborate with cross-functional teams on UI development for testing interfaces and evaluation dashboards
Requirements
- Strong experience with Python, with a focus on building production-grade AI applications
- Proven experience working with LLM frameworks such as LlamaIndex, Haystack, or LangChain
- Experience implementing RAG (Retrieval-Augmented Generation) systems with vector databases
- Solid understanding of AWS services, particularly Bedrock, ECS, S3, and CloudWatch
- Experience with Docker containerization and CI/CD pipelines
- Familiarity with REST API development using FastAPI or similar frameworks
- Knowledge of vector databases (Qdrant, Weaviate, or similar)
- Strong communication skills to interface with both technical and business stakeholders
- BSc or MSc in Computer Science, Data Science, or related STEM field
Desirable Experience
- Experience with prompt engineering and optimization for different LLM models
- Knowledge of advanced RAG techniques like HyDE (Hypothetical Document Embeddings)
- Familiarity with monitoring and evaluation frameworks for LLM applications
- Background in metrology, analytical sciences, or technical documentation processing
- Understanding of embedding models and vector search optimization