Nino Risteski

ML & Data Lead at IVTM Analytics, an open-source contributor, GPU experimentalist and researcher.

I care about building useful and tasteful software, understanding the low level pieces behind deep learning and moving quickly from rough idea to complete product.

Obsessed about compilers, GPUs, ML infrastructure, and AI research.

Lately I’ve been digging into GPU kernels, memory bandwidth, and matrix multiplication performance.

Talks

What The Stack · October 2026

From Portable AI Lab to Frontier Research

Slide from the What The Stack talk on tinygrad, with Nino Risteski presenting

iOSKonf26 · August 2026

From Unified Memory to USB GPUs — Apple-First Deep Learning

Slide from the iOSKonf26 talk on Apple-first deep learning, with Nino Risteski presenting

Projects

Open-source Contributions

Tinygrad

fix same padding stride #19325 (Nov 2026) — fixed stride=(1, 1) handling.
delete extra cast in decompositions #14517 (Feb 2026) — removed a redundant cast.
optim empty shard #13513 (Dec 2025) — optimized empty shard handling.
rope cache optim for jit prune in llm.py (Aug 2025) — cached ROPE angle calculations.
adding enable_gqa in SDPA (Jul 2025) — added grouped query attention support.
clean up unused imports in examples and update CI linting (Jun 2025) — cleaned imports, added CI linting.
fix test logcumsumexp broken devectorize=0 (Jun 2025) — fixed NaN bug in logcumsumexp.
small fix replacing download_file with fetch (Jun 2025) — swapped in the canonical fetch helper.

Apple GPU / Compute

METAL_SGEMM_M4

Metal compute shader experiments for SGEMM (4x4 tiles) that explore Apple GPU optimization and benchmarking.

CUDA / Optimization

CUDA_SGEMM_BUDGET_GPU

SGEMM optimization with CUDA on a budget GTX 1660 GPU.

Agent / RAG

TinyPilot

AI agent that helps you learn TinyGrad and keeps you updated on bounties, featuring RAG with ChromaDB and GPT integration.

Search / LLM App

HubermanAnswers

Ask Dr. Andrew Huberman questions and receive answers directly from his podcast episodes.

Desktop Utility

smart_screenshots

A tool that renames screenshots immediately based on the context in the screenshot.

Writing

RW #9 - How Vector DBs Store 100M Embeddings on One Machine (and Still Search Fast) January 11, 2026 - ML Pills
You Won't Do Boring Work in The (Near) Future January 3, 2026 - Worklog
How I Started Contributing to tinygrad — My First 4 Merged PRs August 11, 2025
Is "Jack of all trades - a master of none" the best mindset for a Data Scientist? January 26, 2023

Useful Stuff

Job Application StrategyGabriel Peters on standing out with personalized demos
Reading & ResearchMason Wang on effective reading and research strategies
School is Not EnoughSimon Sarris on learning through doing and building agency
Neural Networks: Zero to HeroAndrej Karpathy's practical deep learning course
Cursor InterviewLex Fridman's conversation with the founding team of Cursor
Hugging Face CoursesFree ML and NLP learning paths
Machines of Loving GraceDario Amodei on how AI could transform the world for the better
The Techno-Optimist ManifestoMarc Andreessen on technology as a lever for progress