About
Research engineer building safe and aligned AI agents.
I'm an AI Engineer, upskilling in AI safety research and engineering to work on agent alignment and technical AI policy and governance.
I bring 2+ years of AI-native software development and AI engineering, along with an LLB (Hons), Law, to bridge the gap between policy and implementation.
My current focus is legal technical alignment — translating policy recommendations into concrete technical implementations, and contributing to recommendations grounded in technical realities.
Selected Work
- jvdb
Vector database implementation in Python.
- BrilliantCode
Open-source, multi-platform agentic IDE — an AI agent supporting OpenAI and Anthropic models inside an Electron environment, written in TypeScript.
- Brilliant AI Chat
Multi-model AI assistant for frontier models from OpenAI, Anthropic, xAI and DeepSeek — unified context and memory, with built-in image generation, web search and RAG.
- LlamaCloud
Unified inference engine for frontier open-source models, behind an OpenAI-compatible API. React + JavaScript frontend, Python + FastAPI backend.
- Bril AI
AI agent for generating long-form courses with video, audio and chaptered text. HTML/CSS/JS frontend, Python + FastAPI backend; Azure OpenAI (o3, gpt-4o-audio) and Vertex AI for inference.
Jennifer Umoke