Aaron Ontoyin Yin
Electrical & Electronics Eng. Student | AI Engineer/Researcher
Electrical & Electronics Eng. Student
AI Engineer/Researcher
About Me
I am an Artificial Intelligence Engineer and Researcher at Vela Partners, and an undergraduate student in Electrical and Electronics Engineering. My passion lies in data science, machine learning, and especially deep learning, where I enjoy building intelligent systems that solve real-world problems.
I am also actively learning quantum computing and expanding my expertise in robotics. I thrive on exploring new technologies and applying them to innovative projects.
At the core, my journey is driven by curiosity, a love for learning, and a commitment to making a meaningful impact through technology.
Vadmin
Skills
- Deep Learning
- Data Science
- Large Language Models
- Software Engineering
- Robotics
- Electronics & IoT
- Quantum Computing
GPTree Invention
GPTree is a general-purpose artificial intelligence algorithm designed to enhance decision-making through explainable, human-in-the-loop collaboration, eliminating the need for extensive feature engineering. When applied to venture capital (VC) investment analysis, GPTree significantly outperformed random selection by approximately 10x and expert benchmarks by about 3x. Fine-tuned for identifying outlier startups, it achieved an 18% success rate, underscoring its potential to transform early-stage investment strategies and drive superior fund performance.
View Published PaperData Transmission in Networks
Developed a Streamlit-based software tool to analyze and visualize data transmission in network environments, supporting PhD-level research. The application models the flow of information between network nodes and examines the factors influencing transmission efficiency, providing actionable insights for optimizing network performance.
View ProjectDrillbotics Research Competition
As a key member of the UMaT team—Africa’s only representative in the 2023 Drillbotics international competition—I led the development of a Python package for simulating directional drilling. Our team also engineered an intuitive interface for real-time process monitoring, contributing to the team’s recognition at a global level.
View ProjectCustomer Churn Prediction
Designed and implemented a customer churn prediction system utilizing IBM’s telco customer dataset. Leveraged machine learning models including logistic regression, random forest, SVM, KNN, Naive Bayes, and XGBoost. Deployed the top three models via Streamlit, enabling users to prioritize recall, precision, or a balanced approach within the interactive application.
View ProjectAdvanced ML Pipeline for Network Intrusion Detection
Developed a robust machine learning pipeline for detecting network intrusions. Utilized Random Forest for feature extraction and PCA for dimensionality reduction, and implemented an optimized SVM classifier. Integrated the workflow using scikit-learn’s Pipeline object, with comprehensive performance evaluation using metrics such as accuracy, precision, recall, F1 score, ROC curve, and AUC-ROC. Enabled probability estimation and analyzed the impact of decision thresholds on detection performance.
View ProjectOSAD API (Object, Speech and Audio Detection)
Engineered a test-driven, production-ready API for object detection in images and speech-to-text/text-to-speech conversion. The API features secure user authentication, session management, and password reset via email. Built with Flask, it employs TorchVision’s Faster R-CNN model for object detection, and integrates SpeechRecognition and pyttsx for audio processing.
View ProjectUni Hostels App
I develped a web app for easy management and finding of hostels in Universities. This was full scale but just for fun.
View ProjectAcademicRecords
I developed a console based cpp + python program for efficient and easy management of academic records at the university level.
View ProjectEASEchaos
Together with other students, we built an app that extracts the timetable for any class from a general timetable that is always released by our university.
View ProjectElectronics
For electronics, I've built countless projects including Fingerprint + keycard attendance system, automous vehicles, different types of smart entropy minimizers, etc.
View Project- Deep Learning
- Data Science
- Space Exploration
- Quantum Computing
- Watching Sci-fi Movies/Series
- Listening to music
- Playing games (VR mostly)
Research Publications
You can tell a person is eager to be a better version of themselves when they are just ready to learn new skills you introduce them to. I have a trained many people in the Aaenics Community and Aaron is such a person who just needed an initial push. I am amazed at how far he has gone in tech and robotics, especially with Machine learning, and can only imagine where he will be in the next couple of years.
Aaron Ontoyin is a disciplined, organised and an analytical student. He is also a hard working and highly motivated individual. He has a pleasant personality, a team player and a result-oriented person.
I highly recommend Aaron Ontoyin. I had the pleasure of working with him on two significant projects: the drilling simulator automatic software for SPE and an opportunistic routing network project. Aaron's exceptional willingness to learn, adapt, and think critically in conjunction with programming played a pivotal role in the success of both endeavors. His dedication, problem-solving skills, and collaborative spirit make him an invaluable asset to any team. Aaron's commitment to excellence is truly commendable, and I have no doubt he will continue to excel in his future endeavors.
I have known Aaron Ontoyin Yin for some time now. His exceptional skills and talent in programming have made him stand tall among his colleagues. He is very dedicated to his work. His love for computer programming has made me admire him very much. He is very knowledgeable about the Python programming language. Aaron is one of a kind. Contact Aaron for 100% work done.
