Space · AI infrastructure
SpaceVision
A simulator that compares how an AI request travels to a data center on Earth or in orbit.
VisitNanoFID, by Oleh Lahoda
See what
the model sees.
Train a face recognition model with your own photos. Watch what it learns. See where it gets things wrong.
The idea behind itA confident prediction can still be wrong.
Most people only see the result of a face recognition system. NanoFID lets you see the training, test its accuracy, and understand the gap between a score and certainty.
I built it as an independent research project on model behavior, fairness, and consent. The goal is simple: make the system easier to question.
The person behind the project
Oleh Lahoda.
I build software that turns hard ideas into useful tools. My work moves between machine learning, quantum computing, education, and the interfaces that make them understandable.
At Stanford Pre-Collegiate Summer Institutes, I researched Predicting Recurrence of Thyroid Cancer Using Machine Learning and studied multivariable calculus.
About the programI study at Dublin School and have also completed Harvard CS50 and MIT coursework in programming and machine learning.
I'm a software engineering intern at Hatz.AI , and a full-stack developer at Bates Publishing.
Beyond NanoFID.
Quantum computing
QRouter
One API for analyzing and routing quantum circuits to the hardware that fits the job.
VisitEducation software
Bates Math
Daily math practice built from Bates Publishing textbooks, with grading and progress tracking.
VisitI also published Routing Efficiency of AI Data Centers in Space, the engineering paper behind SpaceVision.
Read the paper