I'm Rising Odegua, a seasoned full-stack product and research engineer building frontier world-model platforms, developer APIs, demos, and agent-ready product surfaces.
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 years shipping user-facing B2B and B2C products, and co-created open-source data tools used by thousands of developers.
Strong languages: TypeScript, JavaScript, Python
Focus areas: World models, AI product engineering, developer APIs, model serving, eval tooling, 3D in the browser, open-source data tools
- 🔭 Currently building world-model products, APIs, agent integrations, and browser-native 3D demos at spAItial.
- 🧠 Exploring 3D Gaussian Splatting, browser RL, WebGPU-adjacent workflows, and agent-ready world-model interfaces.
- 🛠️ Previously co-created and maintained open-source data tools like Danfo.js, Datasist, and Scikit.js.
- 👯 Open to collaborating on developer tools, AI product infrastructure, eval tooling, world-model demos, and data/ML open source.
- 💬 Ask me about full-stack AI product engineering, OSS, model-serving workflows, data tooling, and shipping research into products.
- 📫 Connect with me: LinkedIn, Website, Google Scholar.
- Building and launching spAItial's world-model platform, including app.spaitial.ai, the developer API, demos, and agent-ready integrations.
- Built bot-in-world, a browser demo where a humanoid bot learns to navigate worlds generated by the SpAItial API using client-side reinforcement learning.
- Built segment-3gs, a browser-based 3D Gaussian Splatting segmentation playground powered by SpAItial worlds.
- Released Amazing Calendar, an AI calendar assistant with multi-source sync, semantic search, and an agent-based query pipeline.
- Co-authored Building Data-Driven Applications with Danfo.js.
- Became a GitHub Star.
- Became a Google Developer Expert in Machine Learning.
- Danfo.js: pandas-like data library for JavaScript, 5k+ stars, featured by the TensorFlow team.
- bot-in-world: browser RL demo for training a humanoid bot inside SpAItial-generated worlds.
- segment-3gs: real-time 3D Gaussian Splatting segmentation playground.
- Amazing Calendar: AI-powered multi-calendar assistant.
- PortalJS: open-source framework for building rich data portals.
- Datasist: Python library for data analysis, visualization, exploration, and modeling.
- Building an Intelligent Calendar Assistant
- Introducing Danfo.js: a Pandas-like library in JavaScript
- How to put machine learning models into production
- How to Serve Machine Learning Models with TensorFlow Serving and Docker
- Latest AI Engineering Talk
See more technical articles on Medium.
Learn more on my website.