Hands-on engineering bootcamp on production-grade RAG. Hybrid retrieval, reranking, evaluation harnesses, and agent integration.
The AIMenta RAG Engineering Bootcamp is a three-day intensive for hands-on AI engineers and ML platform architects building production retrieval-augmented generation systems. The May 2026 cohort runs in Singapore, with a hybrid participation option for engineers based in other APAC markets.
The curriculum covers the full RAG engineering stack: document ingestion and chunking strategy, embedding model selection and fine-tuning, vector database architecture (HNSW vs. IVF-Flat, filtering strategies), hybrid search design (BM25 + dense retrieval), reranking models, context-window management, hallucination mitigation patterns, evaluation frameworks (RAGAS, DeepEval), and deployment architecture for production latency targets. Day 3 is a capstone build session where participants implement a production-grade RAG pipeline on their own domain corpus.
Prerequisites: proficiency in Python and at least one cloud platform (AWS, GCP, or Azure). Familiarity with LangChain or LlamaIndex is helpful but not required. All tooling runs on participants' own cloud accounts during the workshop. Group size is capped at 16 to maintain a hands-on ratio. Contact AIMenta to secure a seat or to arrange a private in-company edition.