Hi, I'm Gul Hassan
Graduate Engineer at Neural Lab, Singapore — I build and test production ML systems, from deep-learning models to the pipelines and APIs that ship them.
Turning data into decisions, code into impact.
I'm an AI/ML engineer with a strong foundation in deep learning, computer vision, and data science. I hold a BS in Information Technology (GPA 3.35) and have experience building production-ready ML systems, web apps, and automated testing pipelines.
I was selected as a Section Leader for Stanford's Code in Place in 2025 and again for 2026, and have won multiple international hackathons.
Where I've Worked
Professional and volunteer roles that shaped my career.
- Designed and executed regression test suites to validate critical updates for OpsAssist, an airport operations platform in its testing phase.
- Automated regression workflows using Selenium and Appium, leveraging AI-driven testing techniques to optimize efficiency.
- Collaborated with developers to detect, document, and resolve defects, supporting readiness for large-scale operational rollout.
- Transitioned into backend development, managing databases and data processing pipelines on AWS and Huawei Cloud.
- Built and maintained GraphQL APIs, and contributed to an AI-driven model for optimal task assignment deployed to production.
- Selected again as a Section Leader for Stanford's Code in Place 2026.
- Teaching Python to a global cohort and leading technical discussions on algorithms, debugging, and OOP.
- Chosen from 2,000+ global applicants to teach Python to Stanford's largest-ever cohort of 20,000 learners.
- Led technical discussions on algorithms, debugging, and OOP alongside a teaching team spanning 180 countries.
- Contributed to standardizing coding pedagogy and remote learning strategies for large cohorts.
- Taught 4+ courses on data analysis (Google Sheets, Data Analysis 1 & 2) and a 5-week IELTS course.
- Designed and conducted a comprehensive Python course, covering fundamentals to intermediate levels.
- Conducted workshops on Python, Data Analysis, VS Code, Git, and GitHub.
- Managed workshops as a lead on DSA, Computer Vision, Machine Learning, and Exploratory Data Analysis.
- Secured MoU partnerships with 2+ universities, student organizations, and companies.
- Ranked among the top 3 of 20 regional ambassadors, surpassing MoU signing targets by 99%.
- Fostered community growth through strategic partnerships in Pakistan.
- Partnered with international clients on 20+ analytical projects, providing data-driven insights and solutions.
- Developed and maintained interactive dashboards using Google Sheets and Data Studio to track key performance metrics.
- Delivered comprehensive reports that identified trends and supported business decision-making.
Featured Work
AI/ML and full-stack projects I've built.
AI-Based Acoustic Recognition System for Avian Species
Wildlife surveys still rely on manual audio tagging — slow and inconsistent. I built an end-to-end classifier where librosa mel-spectrograms feed a custom 6-layer CNN (16–1024 filters), served through a Flask interface for real-time classification.
Emotion Recognition from EEG Signals
Emotion detection from brain signals is noisy and usually needs hand-crafted features. I trained an LSTM pipeline directly on EEG sequences with an adaptive learning-rate scheduler, then validated it against a confusion matrix and F1-score.
Automated Brain Tumor Detection Using MRI Scans
Early tumor diagnosis is limited by radiologist time. I fine-tuned EfficientNetB0 via transfer learning on a Kaggle MRI dataset with OpenCV preprocessing to classify scans automatically.
Sentiment Classification with Computational Linguistics
Rather than assume a model, I benchmarked five algorithms on an NLP preprocessing pipeline (NLTK + TF-IDF) to find the most reliable text classifier, then shipped the winner as a live app.
Intelligent Handwritten Digit Recognition
I benchmarked a classical SVM against deep nets on normalized, resized MNIST data to test how far simpler models go before reaching for a neural network.
Achievements
Recognition and milestones throughout my journey.
International Hackathons & Competitions
Turning real-world problems into working ML prototypes under tight competition deadlines.
Took part in 12+ international hackathons — competing against teams worldwide and placing in several finals.
World Innovation Day Hackathon
Analyzed disaster data to quantify the human cost of natural disasters and surface actionable trends.
- Processed disaster datasets (floods, tornadoes, landslides) with NumPy, Pandas, Matplotlib, and Seaborn, managing data in Google Sheets.
- Produced reports highlighting trends and how many people were severely and partially affected.
Formula AI Hackathon
Predicted race outcomes from historical Formula AI data using a trained ML model.
- Applied feature engineering, data imputation, and text preprocessing to prepare the race dataset.
- Trained the model to predict outcomes based on weather, track characteristics, and past performance.
World Innovation Day Hackathon
Built an ML model to predict food waste from everyday meal context.
- Used inputs like food name, type, quantity, season, fridge availability, and number of people to predict waste vs. consumption.
- Evaluated with F1-score, Cohen's Kappa, and a confusion matrix to analyze classification effectiveness and error distribution.
Regional Ambassador — World Innovation Day Hackathon
Selected from 250+ applicants as a regional ambassador with Hackmakers (Pakistan).
- Ranked among the top 3 of 20 regional ambassadors, surpassing MoU signing targets by 99%.
- Secured 6+ strategic partnerships between Hackmakers, universities, student groups, and companies.
CS50x Puzzle Day
Won CS50x Puzzle Day, a global logic and problem-solving competition organized by Harvard University.
My Toolkit
Technologies and tools I work with every day.
Machine Learning & AI
Programming & Backend
Data & Analytics
Tools & Platforms
Articles & Notes
I write about machine learning, Python, and problem solving.
Sentiment Analysis hands-on using NLP
An end-to-end walkthrough — from the theory of sentiment analysis and NLP to preprocessing, TF-IDF features, and training an SVM model step by step.
Read on MediumLinear Search Using Python
Why data structures and algorithms matter, and how linear search works — explained with worked examples and accompanying video tutorials.
Read on MediumDictionary in Python
A beginner-friendly guide to Python dictionaries — creating, accessing, updating, and deleting key–value pairs, plus a real-world counting problem.
Read on MediumLet's Work Together
Have a project in mind? Let's make something amazing.
I'm always open to new opportunities, collaborations, and interesting projects. Feel free to reach out!