---
title: "Gaurav Kumar (@gauravbytes) | Hashnode"
description: "Backend Engineer & AI Builder. Read the latest articles by Gaurav Kumar on Hashnode."
url: "https://hashnode.com/@gauravbytes"
---

@gauravbytes

# Gaurav Kumar

@gauravbytes · New Delhi · Joined October 2021

Backend Engineer & AI Builder

[Website](https://gauravkumar-phi.vercel.app/)

### [The 5 AI Engineering Problems That Break Most Production Agents](https://gauravbytes.dev/the-5-ai-engineering-problems-that-break-most-production-agents?utm_source=hashnode&utm_medium=feed)

Explore the five hardest AI engineering problems: context, cost, evals, guardrails, and secure tool access for production agents.

7 min read · [gauravbytes.dev](https://gauravbytes.dev/the-5-ai-engineering-problems-that-break-most-production-agents?utm_source=hashnode&utm_medium=feed)

[Nov 3](https://hashnode.com/posts/the-5-ai-engineering-problems-that-break-most-production-agents/6ae99a854c1e8e8b9de506a0)

### [Why Modern Microservices Need Event-Driven Architecture](https://gauravbytes.dev/why-modern-microservices-need-event-driven-architecture?utm_source=hashnode&utm_medium=feed)

In traditional backend designs, components communicate primarily using synchronous request-response mechanisms (such as REST APIs or gRPC). Service A explicitly calls Service B and blocks execution wh

4 min read · [gauravbytes.dev](https://gauravbytes.dev/why-modern-microservices-need-event-driven-architecture?utm_source=hashnode&utm_medium=feed)

[Aug 31](https://hashnode.com/posts/why-modern-microservices-need-event-driven-architecture/6a953c5ad87132814ffca172)

### [How to Evaluate RAG Retrieval: A Practical Guide to Precision, Recall, MRR, MAP, and NDCG](https://gauravbytes.dev/how-to-evaluate-rag-retrieval-a-practical-guide-to-precision-recall-mrr-map-and-ndcg?utm_source=hashnode&utm_medium=feed)

You implemented a search algorithm in your RAG system and it's returning data without any error, and you think that's it, you're done. You're wrong. That mindset works fine for a normal system — an AP

14 min read · [gauravbytes.dev](https://gauravbytes.dev/how-to-evaluate-rag-retrieval-a-practical-guide-to-precision-recall-mrr-map-and-ndcg?utm_source=hashnode&utm_medium=feed)

[3](https://hashnode.com/posts/how-to-evaluate-rag-retrieval-a-practical-guide-to-precision-recall-mrr-map-and-ndcg/6a70e915ac23f68b04cb65b6) [Aug 3](https://hashnode.com/posts/how-to-evaluate-rag-retrieval-a-practical-guide-to-precision-recall-mrr-map-and-ndcg/6a70e915ac23f68b04cb65b6)

### [RAG Search Algorithms - How Retrieval Actually Works Under the Hood](https://gauravbytes.dev/rag-search-algorithms-how-retrieval-actually-works-under-the-hood?utm_source=hashnode&utm_medium=feed)

Every agentic AI system that "looks things up" before answering — a support bot searching a knowledge base, a coding agent searching a codebase, a research assistant searching PDFs — is running a Retr

13 min read · [gauravbytes.dev](https://gauravbytes.dev/rag-search-algorithms-how-retrieval-actually-works-under-the-hood?utm_source=hashnode&utm_medium=feed)

[1](https://hashnode.com/posts/rag-search-algorithms-how-retrieval-actually-works-under-the-hood/6a5dea666ff4fffb2f99fde2) [Jul 20](https://hashnode.com/posts/rag-search-algorithms-how-retrieval-actually-works-under-the-hood/6a5dea666ff4fffb2f99fde2)

### [pgvector for RAG Search: Schema, Storage, and Retrieval, With Real Results](https://gauravbytes.dev/pgvector-for-rag-search-schema-storage-and-retrieval-with-real-results?utm_source=hashnode&utm_medium=feed)

RAG search has four moving parts: split the document, turn text into vectors, store those vectors somewhere queryable, and pull the right ones back out at query time. This post covers the last two — c

6 min read · [gauravbytes.dev](https://gauravbytes.dev/pgvector-for-rag-search-schema-storage-and-retrieval-with-real-results?utm_source=hashnode&utm_medium=feed)

[Jul 20](https://hashnode.com/posts/pgvector-for-rag-search-schema-storage-and-retrieval-with-real-results/6a5dea1f11a9d58fe8320399)
