Skip to content
View MANVENDRA-github's full-sized avatar
5 followers · 0 following

Achievements

Achievement: Pull Sharkx3Achievement: Pair ExtraordinaireAchievement: YOLOAchievement: Quickdraw

Achievements

Achievement: Pull Sharkx3Achievement: Pair ExtraordinaireAchievement: YOLOAchievement: Quickdraw

Block or report MANVENDRA-github

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.

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
MANVENDRA-github/README.md

profile views followers open to collaboration

👋 About Me

I'm a software engineer who is all-in on AI and agentic systems. I'm fascinated by what large language models make possible, and I love turning that frontier into software that actually ships and holds up in the real world. The agentic edge is what excites me most: systems that can reason, plan, use tools, remember, and get better over time.

  • 🤖 Agentic AI — multi-agent orchestration, AI coding agents, tool-use, and memory, backed by evals that keep them honest.
  • 🧠 LLMs, end-to-end — retrieval, routing, caching, verification, and observability; coaxing reliability out of non-deterministic models.
  • 🏗️ Production-first — I take ideas from prototype to deployed, tested, observable systems, not just demos.
  • 🌱 Always building & learning — chasing new agent frameworks, model capabilities, and developer tooling.

💬 Always happy to talk AI agents, LLMs, and the future of how we build software.


🛠️ Tech Stack

Languages

TypeScript JavaScript Python Java Rust C

AI / LLM & Agents

LangChain LangGraph OpenAI Claude

Frontend

React Next.js Tailwind CSS Astro Vue.js Svelte

Backend & APIs

Node.js Express FastAPI Spring Boot

Databases & DevOps

MongoDB Convex Docker Kubernetes Terraform GitHub Actions OpenTelemetry


📊 GitHub Stats

GitHub stats Top languages
GitHub streak
Contribution activity graph

🚀 Featured Projects

🛡️ Sentinel

A verifying LLM gateway — smart routing, semantic cache, inline output verification, and OpenTelemetry tracing with a live dashboard. Shipped v0.1.0: ~50% cache cost-cut, ~14 ms p99 overhead, 100% PII catch-rate.

TypeScript OpenTelemetry LLM

🧠 sage-agent

A memory-augmented LangGraph agent with a full evaluation harness — long-horizon memory plus reproducible evals.

Python LangGraph

⚔️ pandavas

A five-agent bug/feature fixer with a deterministic quality gate — also shipped as a keyless /pandavas skill for Claude Code & Cursor.

Python LangGraph Claude

✈️ TripGenie-AI

An AI travel planner that turns a single prompt into a complete, structured itinerary.

Next.js Convex LLM

💬 ChatBuzz

A real-time chat + video social app on the MERN stack, powered by Stream.

React Node.js MongoDB Stream

🔎 whats-that-command

An intent-based shell-command reference — describe what you want to do, get the exact command.

JavaScript CLI

📈 Profile Summary

Repos per language Most-committed language
Productive time (IST)

🐍 Watch My Contributions Get Eaten

github contribution snake animation

🤝 Connect With Me

LinkedIn X (Twitter) LeetCode Email Blog

Pinned Loading

  1. agentry agentry Public

    Author your AI coding agents, skills, commands, rules, and MCP servers once — sync them to Claude Code, Cursor, Codex, and OpenCode. Curated, harness-neutral, zero dependencies.

    JavaScript

  2. pandavas pandavas Public

    Five agents, one coding task. Point it at a local repo, describe the bug or feature, and walk away — it researches, writes the change, runs the tests, has a separate agent judge the result, and loo…

    Python

  3. sage-agent sage-agent Public

    A memory-augmented chat agent on a LangGraph ReAct loop — it decides for itself when to recall a memory, save one, search the web, or track a goal. Two evaluation suites score whether those decisio…

    Python

  4. Fathom Fathom Public

    WebGPU observability cinema for LLM-ops telemetry — simulate & render ~1M span-particles at 60fps.

    TypeScript

  5. pitch-pilot pitch-pilot Public

    An autonomous SDR agent that turns a company domain into source-cited cold outreach. Every claim it writes is backed by a first-party fact, judged for faithfulness against that source, and queued f…

    Python

  6. sentinel sentinel Public

    A self-hostable, OpenAI-compatible LLM gateway that catches bad responses before your users do — routes to the cheapest capable model, caches semantically, and verifies every call inline (guardrail…

    TypeScript