CV

Summer School

Columbia University

Machine Learning Summer School

Ph.D.

University of Georgia

Computer Science

• NES Lab (Neuro-Symbolic Computing Lab). Advisor: Dr. John A. Miller.
• Committee: Dr. Jin Sun, Dr. Ninghao Liu. Co-advisor on selected work: Dr. I. Budak Arpinar.
• Dissertation: diffusion-based generative models for probabilistic epidemiological time series forecasting.
• Graduate coursework: Machine Learning, Algorithms, Software Engineering, Data Mining, Trustworthy Machine Learning, Databases, Advanced Robotics, Human-Computer Interaction and Advanced Representation Learning.

Bootcamp

42 Silicon Valley

Software Engineering programme bootcamp

B.Sc.

University of Ilorin

Computer Science

MAAN: Multi-channel Adaptive Attention Network for Probabilistic Time Series Forecasting

SIAM International Conference on Data Mining (SDM26) — accepted

Introduces the Causal Cross-Modal Adapter, operating in two modes: a latent funnel for high-dimensional covariates and a noise-adaptive causal graph for low-dimensional covariates. Joint-window training with selective noising and a masked diffusion loss for stable multi-step forecasting.

Ethical Implications of WannaCry: A Cybersecurity Dilemma

Proceedings of the 19th International Conference on Cyber Warfare and Security (ICCWS), Academic Conferences and Publishing International

With J. Osamor, C. Iwendi, F. Olajide, I. Peter-Osamor, V. Onyenagubom and I. Ayodele. Analyses the ethical responsibilities of government agencies, technology companies and security professionals in handling zero-day vulnerabilities, and proposes ethical frameworks balancing national security against global cybersecurity.

Adaptive Quantile Guidance in Diffusion Models: Multi-Dataset Learning for Pandemic Time Series

IEEE International Conference on Machine Learning and Applications (ICMLA) — accepted

Introduces AQDiff: multi-dataset pre-training of a unified time series diffusion model across CDC influenza-like illness, COVID-19 and RSV data, with adaptive quantile guidance driven by an exponentially weighted residual buffer. Reduces mean absolute error by up to 73.3% against CSDI and PatchTST baselines at under 10% additional computational overhead.

First Place, Google Health AI MedGemma Impact Challenge

Google Health AI

USD $30,000 grand prize. Selected first from over 850 teams and 6,500 entrants worldwide, for EpiCast.

Ambassador Award of Excellence

University of Ilorin Alumni

Annual award recognising distinguished alumni of a university that graduates approximately 12,000 students each year.

Outstanding Graduate Teaching Assistant Award

School of Computing, University of Georgia

Departmental award recognising sustained instructional contribution across ten courses in computer science and machine learning, including Evolutionary Computing, Computer Architecture, Human-Computer Interaction and Theory of Computation.

First Runner-Up, Franklin CREATE! Pitch Competition

University of Georgia

Placed first runner-up among 20 competing teams.

Franklin CREATE! LAUNCH Grant

University of Georgia

USD $7,500 competitively awarded build grant; funded development of Pincel.

AWS "We Power Tech" Grant (Project Alloy)

AWS re:Invent

Selected as one of approximately 50 recipients from approximately 3,000 applicants (under 2%).

University of Ilorin student honours

University of Ilorin

Certificate of Honour, "50 Most Reputable CIS Students" (CISSA); Software Developer of the Year; Leadership Award of Excellence; Programmer of the Year; Woman Tech Maker of the Year.

EpiCast: multimodal AI epidemiological surveillance for ECOWAS

Mobile-first surveillance platform for West African community health workers. MedGemma 4B fine-tuned and quantized to GGUF for fully offline on-device inference at sub-second latency with no cloud dependency. Integrates MedSigLIP for clinical image triage, HeAR for cough-based respiratory classification, and a multilingual NLP pipeline trained on 9,700+ corpus-grounded examples across French, Hausa and Yoruba. MedGemma 27B generates structured WHO IDSR situation reports.

• MedGemma 4B/27B
• MedSigLIP
• Google HeAR
• GGUF on-device
• FastAPI

MyFixam: AI marketplace for artisan services, Nigeria

Founder, MYFIXAM LTD. An AI on-demand marketplace connecting Nigerians with verified artisans — plumbers, electricians, tailors, mechanics and 25+ trades — with payments held safely until the job is confirmed done.

• Founder
• Marketplace
• Nigeria

Pincel: research paper to architecture diagram converter

Two-stage pipeline (PDF parsing and methodology extraction, then structured TikZ/SVG generation) producing publication-quality architecture diagrams. Funded by the Franklin CREATE! LAUNCH grant. Used by over 20 researchers.

• NLP/LLM
• TikZ/SVG
• React

Tsaro: AI-powered community early warning system, northern Nigeria

End-to-end system (Flask backend, React Native frontend) combining fine-tuned Gemma models with SMS-tier delivery for low-connectivity environments.

• Gemma
• Flask
• React Native
• SMS delivery

Omnia: AI-powered animated story generation

Generative pipeline transforming narrative input into animated short films via multimodal LLMs for scene decomposition, character-consistent image synthesis and image-to-video animation.

• Stable Diffusion
• AnimateDiff
• React
• Node.js/Express

Machine Learning Engineer (Intern)

Stripe

• Developed and deployed ML models for secret key leak detection, reducing detection latency from 4.5 days to under one hour and preventing multi-million-dollar annual fraud losses. • Engineered large-scale feature pipelines in Spark and Databricks (500M+ records across Iceberg tables), integrated with Shepherd and Airflow for production workflows. • Expanded detection coverage by onboarding new API-call signals (OAuth, Connect), raising recall on confirmed leaks to 94%.

Graduate Teaching Assistant, School of Computing

University of Georgia

• Currently teaching CSCI 4720E: Computer Architecture (70 students). • Ten courses across computer science and machine learning, including Evolutionary Computing, Human-Computer Interaction, Software Engineering, Theory of Computation, Systems Programming and Numerical Simulations. • Recipient of the Outstanding Graduate Teaching Assistant Award, School of Computing.

Software Developer (Intern)

Orbital Schools

• Led development of dynamic websites on the LAMP stack; integrated PayStack and additional payment APIs.

Google Developer Student Lead, Unilorin

Google Developer Student Leads

Among the first cohort of Developer Student Leads selected in Nigeria. Mentored and taught students across 11 African countries in web and mobile development, reaching over 1,000 university students. Received a Letter of Commendation from the Google Developer Ecosystem Program Manager, 31 July 2019.