Skip to content
View vishrk's full-sized avatar

Block or report vishrk

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
vishrk/README.md

Hi there 👋

I'm Vishnu. I'm a software engineer who likes building things and figuring out how they work.

I've spent most of my career working on distributed systems, AWS and cloud applications. Lately, I've been spending a lot more time with AI, LLMs, RAG, agentic systems and MCP.

I enjoy taking an idea, turning it into a small project, and seeing where it goes. Some of those projects end up being useful. Some are just me satisfying my curiosity.

A few things I'm usually interested in:

  • Building software that can actually survive production
  • Understanding what's happening under the hood, especially with AI systems
  • Playing around with new tools and ideas
  • Travelling and finding places worth going back to

You can find me on LinkedIn or reach me at email.

Pinned Loading

  1. agent-rollback agent-rollback Public

    Undo infrastructure for AI agents using sagas and compensating transactions. Roll back tool calls fully or to any checkpoint.

    Python

  2. cascade-router cascade-router Public

    A complexity-based LLM router that balances answer quality and cost by routing prompts to the cheapest model likely to solve them, with confidence-based escalation and evaluation.

    Python

  3. rag-pipeline-pgvector rag-pipeline-pgvector Public

    RAG pipeline focused on retrieval quality, failure analysis, and evaluation with pgvector and HNSW.

    Python

  4. tool-calling-agent-loop tool-calling-agent-loop Public

    Multi-tool LLM agent loop built from scratch on the Anthropic Messages API, with tool evaluation, async execution, streaming, and context management.

    Python

  5. smart-commit smart-commit Public

    Generate conventional commit messages from your staged git diff using Azure OpenAI.

    JavaScript