Ry Nduma

Engineer, researcher, and artist focused on building computational tools that empower humanity.

About

I currently work at the Walsh Materials Design Research Group and study at Imperial College London, but I was born and raised in Nairobi, Kenya.

I am actively developing extensive experience in materials informatics, LLMs, computer vision, and autonomous robotics alongside my studies, and I am eager to build and collaborate on projects that accelerate materials discovery through AI and robotics.

In my downtime, you'll likely find me hanging out with friends, running, playing Ghost of Tsushima, binging Attack on Titan or immersing myself in a good read.

Oh, and I'm also working on improving my Chinese while 한국어 배우기

Work Experience

View complete experience →

Projects

Featured

Crystalyse

A provenance-enforced scientific agent for computational materials design. Crystalyse orchestrates computational tools through natural language while ensuring every numerical result traces to explicit calculations, addressing the core challenge of AI hallucination in scientific domains.

From a terminal prompt like "Suggest a new Na-ion battery cathode", the system autonomously computes capacity (193 mAh/g) and voltage (3.7 V) in ~90 seconds—despite having no pre-coded battery workflows. The agent reasons about which fundamental calculations to chain together, then derives electrochemical properties from first principles.

The goal is to intelligently and responsibly accelerate the computational parts of materials design so scientists can focus their energy on the challenging discovery questions.

AI Agents Materials Design LLMs Scientific Computing

Sky

An agentic synthesis exploration tool using recursive search algorithms and LLM reasoning to predict synthesis routes for novel materials. Winner of the Global LLM Hackathon for Materials Science & Chemistry.

LLMs Materials Science Agentic AI

MatSX

AI-powered platform for materials similarity search and anomaly detection using Magpie composition descriptors and MACE structure embeddings. Winner of the KRICT ChemDX Hackathon.

AI Materials Informatics 3D Visualization

SignSpeak

A real-time sign language translator leveraging computer vision, neural networks, and large language models to bridge communication between Deaf/hard-of-hearing individuals and hearing people. Features bidirectional translation, skeletal tracking with MediaPipe, and context-aware LLM processing.

Computer Vision Neural Networks LLMs OpenCV

Wordle Solver

Built a small transformer based off learnings from Karpathy's makemore course that learns the statistical patterns and structures of five-letter English words and uses them to solve the wordle in as few tries as possible.

Transformers NLP

MARINA AI

A chatbot built on Tensorflow back in 2018 for submission in the Google Science Fair that would serve as a proof of concept for using AI agents with VR to help those dealing with depression.

TensorFlow Chatbot VR

Minesweeper AI

Using learnings from CS50 AI course, I built a minesweeper game that relies on knowledge engineering and has an agent safely navigate a minesweeper game without tripping a mine.

Knowledge Engineering Game AI

Unbeatable Tic-Tac-Toe

Built an AI agent that uses a minimax algorithm to find the most optimal counter to the move made by the player while optimising for its worst case (a tie - no losses!).

Minimax Game Theory Python

I swear the list of side projects keeps growing by the days - some are still in the pipeline being built (in hidden repos for now) but here are some I am proud of.

View more on GitHub →

Publications

Awards & Honors

Community & Volunteer Work

Writing

For half a decade, I've been spilling my thoughts across the digital cosmos (or as my mom calls it, "that blog thing you do").

While it all began with a powerful urge to create scientific awareness across the board – it has beautifully morphed into something far more personal.

Today, my little Substack, Ry's Meta, serves as my digital diary, creative playground, and intellectual sandbox - because in a way it helps me obtain a clarity of thinking that I'm learning can sometimes only come about by writing.

Contact

This site is open 24 hours a day, 7 days a week. Thanks for visiting!

Oh! Before I go, here's a playlist you might enjoy!

LinkedIn GitHub Twitter Email