Skip to content
View SupaTx's full-sized avatar
🎯
Focusing
🎯
Focusing
  • Nigeria
  • 12:19 (UTC +01:00)

Block or report SupaTx

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
SupaTx/README.md

Elijah Olalere SupaTX

AI Engineer · Machine Learning Specialist

Building intelligent systems that bridge research and production.


Python TypeScript React Native Next.js LightGBM Status


About

I'm an AI Engineer based in Nigeria, specialising in machine learning systems from model architecture and training to production deployment and agent design. I work across the full stack, using Python for everything ML and TypeScript with React Native and Next.js on the frontend.

My work sits at the intersection of predictive modeling, LLM agent systems, and real-world deployment. I don't just build models that work in notebooks I build systems that work in production.

Currently building PredictaX9, a mathematically consistent AI prediction engine for the 2026 FIFA World Cup, entering the Stair AI World Cup Agent Arena a live competition where AI agents place real prediction market bets and are scored on both profit and reasoning quality.


Tech Stack

Languages Python · TypeScript
ML & AI LightGBM · Scikit-learn · Scipy · Poisson Modeling · LLM Fine-tuning (PEFT, LoRA)
LLM & Agents IBM Granite · LFM 2.5 · Tool Calling · Agentic Pipelines
Frontend React Native · Next.js
Specialisation Predictive Modeling · ML Inference APIs · AI Agent Design · Sports Analytics

Featured Project

AI-powered football prediction engine for the 2026 FIFA World Cup

  • Trained on 49,000+ international matches from 1872 to 2026
  • Uses Poisson regression to predict expected goals per team
  • Uses multiclass softmax for match outcome — probabilities always sum to 100%
  • All Over/Under, BTTS and Clean Sheet markets derived from pure math — zero contradictions
  • Built as a competing agent in the Stair AI World Cup Agent Arena ($4,500+ prize pool)

What I'm Working On

  • 🏆 Competing in the Stair AI World Cup Agent Arena building an AI agent that makes live prediction market bets on 2026 FIFA World Cup matches
  • 🤖 Wrapping SupaTX Oracle with an LLM agentic layer using LFM 2.5 1.2B Thinking for live data fetching and reasoning
  • 🔧 Fine-tuning LLMs for domain-specific tasks using PEFT and LoRA
  • 📱 Shipping cross-platform AI-powered apps with React Native and Next.js

GitHub Stats

SupaTX's GitHub Stats

Top Languages


Get In Touch

📬 supatxr@gmail.com  ·  🐙 GitHub  ·  💼 LinkedIn


Based in Nigeria 🇳🇬 · Open to interesting problems, collaborations, and remote opportunities.

Popular repositories Loading

  1. TheDataAthlete TheDataAthlete Public

    Football betting prediction model using LightGBM. Predicts total goals, Over/Under 0.5–4.5, 1X2, BTTS, and clean sheets. Trained on 2000‑2025 match data + Elo ratings.

    Jupyter Notebook 2

  2. SupaTx SupaTx Public

  3. PredictaX-9 PredictaX-9 Public

    ⚽ AI-powered football prediction engine for the 2026 FIFA World Cup Poisson regression + multiclass outcome modeling. Built for the Stair AI Agent Arena.

    Python