AI & ML
When Not to Use Similarity Search in a RAG Pipeline
When we learn about RAG, the usual flow feels very natural. A user asks a question, we create an embedding, search the vector database…
NCNavindu Chathuranga · · 5 min read
Writer6 articles
Navindu Chathuranga
Software Engineer @ Enlear
AI & ML
What If You Could Control Every AI Call From One Place?
What If You Could Control Every AI Call From One Place? A practical guide to putting one layer between your application and AI models, so you can track costs, switch models, manage access, and …
Navindu Chathuranga · · 14 min read
AI & ML
Beyond Similarity Search: Metadata Filtering for RAG with Amazon S3 Vectors
Beyond Similarity Search: Metadata Filtering for RAG with Amazon S3 Vectors Semantic search is one of the most useful parts of RAG. A user does not need to ask a question using the exact same words …
Navindu Chathuranga · · 6 min read
AI & ML
Building a Structure-Aware RAG Chunking Pipeline with Amazon Bedrock and S3 Vectors
Building a Structure-Aware RAG Chunking Pipeline with Amazon Bedrock and S3 Vectors A RAG system can use a good embedding model and a reliable vector database, but the results can still feel less …
Navindu Chathuranga · · 10 min read
AI & ML
Building a Serverless RAG Pipeline with Amazon S3 Vectors and Amazon Bedrock
Building a Serverless RAG Pipeline with Amazon S3 Vectors and Amazon Bedrock When I first learned about Retrieval-Augmented Generation (RAG), the architecture looked simple: Store some documents → …
Navindu · · 8 min read
AI & ML
Prompt Injection in AI-Powered Applications
Prompt Injection in AI-Powered Applications How to Test Against Prompt Injection in AI-Powered Applications? Introduction Every AI system operates on instructions. But what happens when someone slips …
Navindu · · 8 min read