Skip to content
View sujee's full-sized avatar

Highlights

  • Pro

Block or report sujee

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
sujee/README.md

Hi, I'm Sujee 👋

I’m an AI Developer Advocate at Nebius, helping developers build with open models through technical content, practical examples, guides, demos, and hands-on workshops.

I bring 25+ years of software-engineering experience across distributed systems, data engineering, machine learning, and cloud infrastructure.

🔭 What I’m working on and exploring

  • 💻 Coding with open models — evaluating open coding models with tools such as Claude Code, Codex, OpenCode, Cline, and Cursor
  • 🤖 AI agents and agentic workflows
  • 🔎 Retrieval-augmented generation, embeddings, and vector databases
  • ⚡ LLM inference, performance, and AI infrastructure
  • 🧹 Data preparation and document processing for AI
  • 🧪 Evaluation of model and agent-harness combinations
  • 🦾 Physical AI and robotics
  • 🧑‍🏫 Hands-on workshops, demos, and developer education

🧰 Technology stack

Models and inference: Open models, Ollama, Nebius Token Factory, hosted and self-hosted inference

AI and agent frameworks: LangChain, Deep Agents, Tavily, OpenWiki, Docling

Data and retrieval: RAG, embeddings, Milvus, vector databases

Coding agents: Claude Code, Codex, OpenCode, Cline, Cursor

Languages and infrastructure: Python, Docker, Kubernetes, cloud infrastructure

💻 Coding with open models

I’m exploring how open models perform in practical, day-to-day software-engineering workflows—not just isolated benchmarks.

I’m particularly interested in repository-scale code understanding, feature implementation, debugging, refactoring, tool use, long-horizon coding tasks, and cost per task.

Models I’m currently testing: Kimi K2.7 Code, GLM-5.2, and MiniMax-M3.

🛠️ Projects I’m building

A compatibility proxy for connecting Claude Code and other coding agents to open-model inference services such as Nebius Token Factory.

A full-stack, open-source RAG chatbot that answers questions about your website using open models, document processing, embeddings, and vector databases.

A hands-on collection of experiments covering LLMs, agents, RAG, inference engineering, and open-model workflows.

Practical examples using Data Prep Kit, showcasing document processing, data preparation, Docling, Milvus, and open-source RAG systems.

🎤 Talks and workshops

I speak and teach about:

  • Coding agents and open models
  • AI agents and agentic systems
  • RAG and vector databases
  • LLM inference and infrastructure
  • Open-source AI tools and workflows

🔗 Connect with me

🌐 sujee.dev • 🐙 GitHub • 💼 LinkedIn • 🐦 X • 🦋 Bluesky • 🎥 YouTube • 💬 Discord: @sujee.dev

Pinned Loading

  1. llm-snake-arena llm-snake-arena Public

    🐍 A live visual arena where LLMs compete in Snake - compare model latency, reliability and decision-making across OpenAI-compatible APIs.

    JavaScript

  2. ai-playground ai-playground Public

    my code and experiments in AI

    Jupyter Notebook 1 2

  3. data-prep-kit-examples data-prep-kit-examples Public

    Examples of using IBM data prep kit

    HTML 32 10

  4. The-AI-Alliance/allycat The-AI-Alliance/allycat Public

    Chat with your website using LLMs

    Python 80 38