Skip to content
View Manitchahar's full-sized avatar
🎯
Focusing
🎯
Focusing
1 follower · 4 following

Achievements

Achievement: YOLOAchievement: Pull Shark

Achievements

Achievement: YOLOAchievement: Pull Shark

Highlights

  • Pro

Block or report Manitchahar

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
Manitchahar/README.md

Manit Kumar

GenAI / LLM Engineer building practical AI systems with Python, TypeScript, RAG, AI agents, ServiceNow automation, and cloud AI platforms.

I like turning rough AI ideas into working tools: agentic workflows, retrieval systems, document intelligence, CLI operators, and automation that helps support and engineering teams move faster.

What I Work On

  • LLM applications with RAG, semantic search, tool calling, and agent orchestration
  • Python/FastAPI services, TypeScript tooling, Telegram/CLI automation, and production-ready APIs
  • ServiceNow and ITSM automation, incident analytics, SOP generation, and operational intelligence
  • OCR/document AI workflows, including human-in-the-loop review for sensitive outputs
  • Cloud AI experiments across Azure OpenAI, Google Vertex AI, AWS, and local model tooling

Featured Projects

Project What it shows
gemini-claw Telegram-native Gemini CLI operator with private allowlisted chats, task workers, session continuity, stream-JSON parsing, and safety controls.
RocketGuard-1B Fine-tuned MiniCPM5-1B model for guardrail and agent safety decisions (allow/block/rewrite/ask). Weights on Hugging Face.
ADK-Agents Google ADK multi-agent prototype with background workers, MCP tools, memory hooks, and eval scaffolding.
bloom AI Content Marketing SaaS — multi-format text generation, image generation, and content improvement (48-hour challenge build).
thousand-token-quest Retro Gradio RPG that turns your resume into a playable career adventure (Build Small Hackathon).
GLM-OCR contribution Merged open-source fix to a 6k+ star OCR project: layout config id2label support, safer missing mapping behavior, and unit tests.

Note: This profile is a public timeline — everything stays visible. Some repos are polished products, most are learning experiments. The featured projects above are the ones worth clicking first.

Core Stack

  • Languages: Python, TypeScript, JavaScript, SQL
  • AI/LLM: RAG, agents, MCP, LangChain, embeddings, vector search, prompt engineering, OCR/document AI
  • Backend: FastAPI, REST APIs, Pydantic, Node.js, Docker, PostgreSQL/pgvector
  • Cloud/Platforms: Azure OpenAI, Azure AI Foundry, Google Vertex AI, AWS EC2/S3, ServiceNow
  • Data/ML: BERTopic, clustering, semantic search, ChromaDB, FAISS, notebooks, operational analytics

Current Focus

  • Turning useful experiments into cleaner projects with stronger READMEs, tests, and reproducible setup
  • Improving practical AI workflows for incident intelligence, document processing, and developer automation
  • Contributing small, useful fixes to open-source AI/OCR tooling

Open Source

  • Merged contributor to zai-org/GLM-OCR, improving layout configuration handling and test coverage.

Reach Me

Focused on useful AI systems, clean developer workflows, and practical automation.

Popular repositories Loading

  1. gemini-claw gemini-claw Public

    Telegram-native Gemini CLI operator with private allowlisted chats, background workers, session continuity, and stream-JSON parsing.

    TypeScript 11 2

  2. llmops-chatgpt-huggingface llmops-chatgpt-huggingface Public

    Forked from manifoldailearning/llmops-chatgpt-huggingface

    LLM operations experiments with ChatGPT and Hugging Face models.

    Jupyter Notebook

  3. LLama-rag-chat LLama-rag-chat Public

    Retrieval-augmented chat experiments with local LLM-backed document search and embeddings.

    Python

  4. chatbot chatbot Public

    Python chatbot experiments with Llama-style local LLM workflows.

    Python

  5. FastTweet FastTweet Public

    Python experiments around fast tweet and post generation workflows.

    Python

  6. EduGenie EduGenie Public

    AI tutor experiments with personalized lessons, quiz generation, and RAG-assisted learning.

    Jupyter Notebook