Profile

Williams Odunayo

Machine Learning Engineer

"Always Learning, Always Building."

Building real, usable AI and ML systems that work — practical, deployable, and impactful.

About Me

Who I Am

I got into tech because I was curious about one thing — how machines learn from data and start to understand the world around us. That curiosity pulled me into computer vision, model training, and building AI that actually works.

Now, I focus on creating systems that are not just smart, but practical — models that can run efficiently, be deployed easily, and make real impact where it matters.

Going forward, I’m exploring more autonomous, agent-like AI systems — ones that can make decisions, adapt, and collaborate. My philosophy stays simple: learn fast, build what works, and keep pushing what AI can do.

What I Focus On

Machine Learning

Designing and deploying ML systems that solve real-world problems — from classification to anomaly detection.

Deep Learning

Working with CNNs, RNNs, and Transformers for tasks like medical imaging, emotion detection, and NLP.

Data Engineering

Preprocessing, feature engineering, and creating scalable data pipelines for model training and evaluation.

Cloud & MLOps

Deploying models using Flask, Docker, and Render. Familiar with model versioning and optimization for production.

Tools & Frameworks

Languages

  • Python
  • JavaScript (basic familiarity)

ML & DL

  • TensorFlow
  • PyTorch
  • Scikit-learn

Computer Vision

  • OpenCV
  • Mediapipe
  • CNNs

NLP

  • Transformers
  • LangChain
  • Gemini API

Key Projects

BrainScan AI

BrainScan AI – Brain Tumor Detection

MRI-based brain tumor classification using TensorFlow & OpenCV. Achieved over 90% accuracy; deployed on Render via Flask + TFLite.

TensorFlow Flask TFLite OpenCV
View Demo
EmotiSense

EmotiSense – AI Emotion Detection

Real-time facial emotion recognition using CNNs and Mediapipe. Supports both .h5 and .tflite models with optional voice feedback.

TensorFlow OpenCV Flask Mediapipe
View Demo
Ask Your Doc

Ask Your Doc – AI Document Chatbot

Lightweight chatbot that lets users "chat" with their documents (PDF, text). Built with Flask, LangChain, Gemini API, and FAISS for embeddings.

LangChain Flask Gemini API FAISS
View Demo

Interests & Future Direction

Agentic AI Systems
Reinforcement Learning
Computer Vision
Production-ready AI
AI Applications in Healthcare & Cybersecurity
Generative AI
Geospatial Intelligence
Open-source Projects
Real-World ML Systems

Get In Touch

Connect With Me

Send Me a Message