Skip to content
View omartood's full-sized avatar
👋
One of my most productive days was throwing away 1000 lines of code👨‍💻👩‍💻
👋
One of my most productive days was throwing away 1000 lines of code👨‍💻👩‍💻

Organizations

@goobolabs @dugsiiyeinc @soplang

Block or report omartood

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
omartood/README.md
Omar Tood — MSc in Artificial Intelligence · AI Researcher · LLM Engineer
Focus: LLMs · Transformers · RAG · AI Agents · Deep Learning · Somali NLP

◇ About

I'm Omar Tood — an AI Researcher and Software Engineer with an MSc in Artificial Intelligence. I build intelligent systems around Large Language Models, agentic AI, and transformer architectures, with a focus on bringing modern NLP to low-resource languages like Somali.


◇ The way I think about AI systems

Pipeline: Dataset → Embedding → Self-Attention → Reasoning → Output

◇ Tech Stack

Python TypeScript   PyTorch Hugging Face LangChain Ollama   FastAPI Next.js PostgreSQL Docker


◇ Featured Projects

🧠 Somali LLM

A large language model built for the Somali language — tokenizer, pretraining, and fine-tuning for quality generation in an underserved language.

🤖 AI Learning Companion

An agentic tutor for AI & data science — retrieval-grounded answers and step-by-step reasoning powered by LLMs.

⚖️ Dastuur AI

A legal-reasoning assistant over constitutional text — retrieval-augmented, citation-aware answers grounded in sources.

📊 Somali AI Benchmark

An evaluation suite measuring how well models understand and generate Somali — standardized tasks and metrics.


◇ Self-Attention, in Somali

Self-attention visualized over a Somali sentence — the kind of low-resource NLP I work on. Each query token lights up and weighs the rest of the sequence.

Self-attention arcs over the Somali sentence: Aniga waxaan baranayaa AI iyo Somali NLP — 'I am learning AI and Somali NLP'

◇ GitHub Stats

GitHub stats Top languages

Training models. Building intelligence. Shaping the future.
Profile views

Pinned Loading

  1. soplang/soplang soplang/soplang Public

    The Somali Programming Language.

    Rust 159 38

  2. goobolabs/sep-ds-ml-bootcamp-2025 goobolabs/sep-ds-ml-bootcamp-2025 Public

    Data Science and Machine Learning Bootcamp. (Sep - 2025)

    Jupyter Notebook 71 65

  3. sharafdin/yonode sharafdin/yonode Public archive

    The Node.js Toolkit for Rapid Development.

    JavaScript 123 13

  4. sophone sophone Public

    Somalia (+252) phone validator, formatter, and operator guesser. Library + CLI.

    JavaScript 12 1