I’m good at two things: explaining machine learning clearly, and spotting where it could actually be useful. Grounded in fraud detection, signal processing, embeddings, and recommendation systems, and explained the same way I write here — from intuition, through the math, to why it works.
What I can help with
- Tailored technical explanations and presentations — including topics I don’t have articles on yet. I’m happy to learn something new and shape it around what your audience actually needs.
- ML opportunity scouting — most companies already know they want to use AI, but beyond chatbots, they’re not sure where. I look at how something is currently done and suggest concrete, creative places classical ML (embeddings, transformers, and similar) could actually fit. This is exploratory: ideas and directions, not a guarantee they’ll work or a full implementation.
- Guest technical writing, or reviewing ML explanations for correctness
- Open to discussing full-time applied scientist roles
See how I think and explain
A few articles that show my approach in practice:
- A brain-friendly guide to PCA: math, visuals, code
- Deriving the SVD from first principles
- Understanding single-head attention in transformers
Availability
I currently work full-time in applied ML, so I’m
taking on a limited number of consulting conversations at a time.
Interested in working together?