At spAItial, my work turns frontier world models into shipped products: app.spaitial.ai, the developer API, model-serving infrastructure, eval tooling, demos, and agent-ready integrations for LLM skills, MCPs, and Claude plugins. Before frontier AI, I spent six years shipping user-facing products (B2B and B2C) and co-created Danfo.js (5k+ GitHub stars, featured by the TensorFlow team) and other open-source tools used by thousands of developers.
<EXPERIENCE/>
{> Member of Technical Staff (Product Engineering)
spAItialSole and leading product engineer at a frontier world-model lab, owning every surface between the models and the people using them. Designed, built, and launched app.spaitial.ai, the developer portal, the public API, and the docs from scratch on new scalable infrastructure, shipped alongside the Echo-2 model release. Made the platform agent-ready with an MCP server, LLM skills, and Claude plugins. Built the eval tooling and the large-scale data sourcing, scraping, and training-data pipelines behind the models. Forward-deployed on customers and partnerships: integrations, launch campaigns, developer relations, and social.
{> Senior Full Stack Software Engineer
KallikorBuilt AI-powered simulation products for turning real-world supply-chain operations into digital twins, giving enterprise teams a way to test decisions in software before changing physical systems.
{> Senior Full Stack Software Engineer
Nossa DataFounding engineer who took Nossa Data's ESG reporting and data-management platform from MVP to production, shaping the architecture, data workflows, and core customer-facing features.
{> Engineering Lead
PhilanthroLabLed development of the Social Safety Net, a search engine for social services powered by a machine-learning-ready knowledge graph that matched people in need with relevant resources.
{> Software Engineer
DatopianCore contributor to PortalJS (2k+ GitHub stars), an open-source framework for building rich data portals. Extended the framework, improved backend integrations including CKAN, and helped teams publish more usable data products.
{> Machine Learning Engineer
Data Science NigeriaJoined as a data science intern and stayed on as a machine learning engineer, working across the full data loop: collection, exploration, transformation, modeling, evaluation, and business-facing insight. Ran workshops and masterclasses on neural networks, and later returned to the DSN bootcamp as a mentor.
{> Software Engineer
Infotech Risks Security
First professional engineering role, taken while still an undergraduate. Built and maintained internal web and mobile tools for a risk and security consultancy.
<SKILLS/>
> Here are my technical skills and areas of expertise:
> Technical Stack
> Domain Expertise
<SELECTED PROJECTS/>
A browser-based 3D Gaussian-splat segmentation playground powered by spAItial worlds. Select and isolate objects inside reconstructed 3D scenes, rendered in real time in the browser.
A browser-based demo where a humanoid bot learns to navigate and fetch inside worlds generated by the SpAItial API. Reinforcement learning runs client-side in a web worker, with PlayCanvas rendering, Rapier physics, generated splats, collision meshes, and pretrained policies.
An AI messaging app that connects every inbox in one place, surfaces the 5% of messages that actually matter, and drafts replies in your own voice so you can act on them in a tap. Launched to a waitlist of 150+ professionals.
A generative-AI mobile app for personalised learning, shipped to the App Store. Turned any topic into bite-sized lessons with a podcast mode for listening on the move and full offline support. Ran for a few months in 2025 before being shut down.
<RESEARCH PAPERS/>
> Optimizing Health Facilities Allocation for COVID-19 Management Using Social Vulnerability Index and Spatial Data Analysis
This study recognises that building new health centers would be slow and expensive in preparation for the pandemic and as such uses social vulnerability index, demographic and environmental statistics to propose suitable existing centers that needs to be re-equipped.
Read Paper >> DataSist: A Python-based library for easy data analysis, visualization and modeling
This paper presents a new python-based library, DataSist, which offers high level, intuitive and easy to use functions, and methods that helps data scientists/analyst to quickly analyze, mine and visualize big data sets
Read Paper >> Predicting Bank Loan Default with Extreme Gradient Boosting
This paper provides an effective basis for loan credit approval in order to identify risky customers from a large number of loan applications using predictive modeling.
Read Paper ><TALKS/>
<BLOGS/>
> Building an Intelligent Calendar Assistant
Exploring the development of an AI-powered calendar app that transforms how users interact with their schedules through natural language and multi-calendar integration.
Read Article >> How to put machine learning models into production
An opinion piece curating best guide in putting machine learning models in production
Read Article >> How to Serve Machine Learning Models with TensorFlow Serving and Docker
Learn how efficiently serve machine learning models with Tensorflow serving
Read Article >> Deep Dive into ML Models in Production Using Tensorflow Extended (TFX) and Kubeflow
An end-to-end tutorial that shows you how to deploy, scale and monitor deep learning models with Tensorflow Extended and Kubeflow on GCP
Read Article ><CONNECT/>
> echo "Let's connect!. Here's how you can reach me:"
> echo "I'm always open to interesting conversations and collaborations!"