Skip to content
View mshojaei77's full-sized avatar
🎯
Focusing
🎯
Focusing

Block or report mshojaei77

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.

Please don't include any personal information such as legal names or email addresses. Maximum 100 characters, markdown supported. This note will be visible to only you.
Report abuse

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

Report abuse
mshojaei77/README.md

Hello there, I'm Mohammad 👋

Machine Learning Engineer | LLM Specialist

Transforming AI research into production systems that drive business value.

Email LinkedIn GitHub Hugging Face Portfolio X

I’m Mohammad Shojaei, a Machine Learning Engineer with a particular focus on Large Language Models and Natural Language Processing. My professional interests revolve around translating cutting-edge research into robust, real-world systems. I spend considerable time refining training pipelines, optimizing inference for low latency, and fine-tuning models to effectively handle multiple languages. The field is dynamic and often unpredictable, but my central objective remains consistent: transforming theoretical concepts into production-grade tools that provide tangible value.

Currently, my primary work involves evaluating the latest open-source LLMs, with an emphasis on those touting comprehensive support for Persian. Keeping pace with the influx of new research is essential to ensuring my systems retain a competitive advantage.

  • My projects include architecting agentic systems with advanced reasoning capabilities, developing conversational audio agents resilient to unexpected queries, and contributing to open-source initiatives aimed at advancing AI accessibility for underrepresented languages.

  • In terms of professional development, I am consistently refining my understanding of AI agent design and evaluation, delving into technical aspects such as tool-calling and structured outputs, and exploring recent post-training methods to enhance the performance of models under 10B parameters.

  • If you wish to discuss advances in NLP, debate the rationale behind embedding dimensions, or exchange insights on deploying models to production environments, I welcome the opportunity for collaboration and knowledge sharing.

🛠️ My Go-To Tech Stack

Languages & Core Libraries:

python pytorch numpy pandas

LLM & NLP Ecosystem:

Langchain/LangGraph transformers ollama

Frameworks & Tools:

fastapi docker git nextjs streamlit

mshojaei77's GitHub Stats

Thanks for dropping by! Let's connect and whip up something awesome.

Pinned Loading

  1. open-notebook open-notebook Public

    AI-powered knowledge management and question-answering system

    Python 36 1

  2. ollama_rag ollama_rag Public

    fully local RAG system using ollama and faiss

    Python 44 5

  3. EasyMCP EasyMCP Public

    A beginner-friendly client for the MCP (Model Context Protocol). Connect to SSE, NPX, and UV servers, and integrate with OpenAI for dynamic tool interactions. Perfect for exploring server connectio…

    Python 16 2

  4. ReActMCP ReActMCP Public

    ReActMCP is a reactive MCP client that empowers AI assistants to instantly respond with real-time, Markdown-formatted web search insights powered by the Exa API.

    Python 140 23

  5. ollama-desktop ollama-desktop Public

    A powerful desktop application for interacting with local AI models through Ollama, featuring advanced capabilities and a modern interface.

    TypeScript 6 1

  6. ParsiEval ParsiEval Public

    a comprehensive benchmark designed to evaluate the performance of Large Language Models (LLMs) on Persian language tasks

    Python 35 1