/ Rajesh Sitaraman - 👨🏻‍💻 ai architect AWS LLM RAG agentic ai building production grade gen ai solutions 2026-04-01T21:22:13+00:00 Rajesh Sitaraman / Jekyll © 2026 Rajesh Sitaraman /assets/img/favicons/favicon.ico /assets/img/favicons/favicon-96x96.png 2026-04-01T00:00:00+00:00 2026-04-01T00:00:00+00:00 /posts/bedrock-model-benchmark/ Rajesh Sitaraman

disclaimer: This post focuses strictly on inference latency, not model quality, accuracy, or reasoning performance. A model that responds in 400ms is not “better” than one that responds in 2 seconds. It is simply faster to start streaming tokens. The right model for your use case depends on many factors beyond latency. This benchmark measures one thing: how quickly each model delivers its fi...

2026-02-15T00:00:00+00:00 2026-03-22T19:55:21+00:00 /posts/the-curious-case-of-sse/ Rajesh Sitaraman

This article was originally published on the AWS Builder Center. The Setup We built a service that runs LLM inference on AWS and streams results back to clients in real-time using Server-Sent Events (SSE). Think of any chat-style LLM application the user submits a prompt, and tokens appear progressively as the model generates them, rather than waiting for the entire response to complete. ...

2025-09-12T00:00:00+00:00 2026-03-17T17:34:47+00:00 /posts/simple-guide-for-rag-using-strands/ Rajesh Sitaraman

This article was originally published on the AWS Builder Center. Intro This guide demonstrates how to build a RAG (Retrieval-Augmented Generation) system using Strands Agents and Amazon’s latest S3 Vectors service. You’ll learn to set up AWS Bedrock Knowledge Base, create vector storage for document embedding, and implement metadata filtering for precise information retrieval. We’ll walk t...

2025-03-31T00:00:00+00:00 2026-03-22T19:56:19+00:00 /posts/browser-use-agent-with-amazon-bedrock/ Rajesh Sitaraman

This article was originally published on the AWS Builder Center. AI-powered automation is changing how we interact with the web, but most solutions are either too complex or too rigid. Meet Browser Use, a simple yet powerful tool that lets AI agents seamlessly interact with websites. It extracts HTML, navigates elements, and supports multi-step workflows all with an intuitive API. Now, im...

2023-01-09T00:00:00+00:00 2026-03-15T04:41:43+00:00 /posts/aws-cdk-custom-resource/ Rajesh Sitaraman

“The information provided in this blog post is generated by ChatGPT, a machine learning model trained on a dataset of text. The views and opinions expressed in this post are solely those of the model and do not reflect the views of the creators or any individual associated with the training of the model. The information provided in this post is for general informational purposes only and is ...