O. Isaac Salako

Hi, my name is Isaac.

I'm a full-stack data scientist and AI engineer who builds end-to-end ML products - from data pipelines and modeling to APIs and deployment.

4+

Years Experience

10+

AI Projects

5

Data Stories

3

Industries Served

Portfolio

UK Housing — Price Trends Explorer

An interactive web app that explores UK house price trends over time and across regions, making it easy to compare areas and spot patterns in the market.

Launch Application

Predicting the Aerodynamic Coefficients of Transonic Airfoils

A machine learning project (undergraduate thesis) that predicts key aerodynamic coefficients (e.g., lift/drag characteristics) for transonic airfoils, with code + experiments and an accompanying published paper.

DOI: 10.30772/qjes.v16i2.955

View on GitHub

DeepInsight — Twitter Sentiment Analysis for the 2023 Nigerian Election

A tweet-driven sentiment analysis system built to track and analyse public sentiment toward different candidates during the 2023 Nigerian election.

View on GitHub

Jun 2025 - Aug 2025

Summer Data Scientist

NIHR BioResource, Cambridge, UK

Built an interactive Plotly Dash sample-tracking dashboard covering 300+ Oxford Nanopore long-read genomic samples/week across pipeline stages, surfacing QC metrics (read length, N50, error rate).
Extended the SQLite schema (3 new metadata/QC tables) and added drill-down filters (by run/date/status/sample ID) to help assistants and bioinformaticians pinpoint bottlenecks in near-real time.

Jan 2025 - Apr 2025

AI Engineer (MSc Advanced Practice)

PocketSergeant, Sunderland, UK

Designed generative-AI pipelines to auto-produce tailored victim-statement checklist templates by synthesising legal/procedural data and cut manual prep time by ~30%.
Prototyped a scenario-based testing API with GPT-4o to create realistic incident narratives.
Shipped REST APIs on Azure App Service supporting ~200 inference requests/day at ~100 ms median latency and ~99% uptime; integrated with mobile & web clients.

Jun 2022 - Apr 2024

Data Scientist

Omeife Technologies, Abuja, Nigeria

Optimised text-to-speech inference, cutting latency by ~40% (500 ms to 300 ms/request).
Engineered features on an 800k-record dataset, improving model accuracy by 7%.
Built end-to-end ML solutions for two UNICCON web apps serving ~1,500 daily users; co-launched ML features and set up continuous model-monitoring workflows.
Developed subtitle-synchronisation algorithms for 100+ hours of video, saving ~5 hrs/week of manual editing.

Experience

Full-stack Data Scientist with 3+ years' expertise in end-to-end model development and deployment. Proficient in Python (pandas, NumPy, scikit-learn), SQL, and cloud services (GCP, Azure) for feature engineering, statistical analysis, and predictive modeling. I am skilled at training, fine-tuning, and evaluating deep learning architectures in TensorFlow and PyTorch, and developing Generative AI assistants for automated technical support.

Education

May 2024 - Present

Teesside University, Middlesbrough

MSc Applied Artificial Intelligence with Advanced Practice

Relevant: ML, Deep Learning, NLP, Data Engineering, Responsible/Explainable AI, Cloud & MLOps.

Jun 2016 - Nov 2021

Kwara State University, Nigeria

B.Eng. Aeronautical and Astronautical Engineering

Relevant: Numerical analysis, computational mathematics, programming, fluid simulation, engineering ethics and safety.