AI by Hand ✍️ Seminar: Frontier - Qwen 3.6
AI by Hand Seminars are back! 🙌
This time, I’ll cover four frontier open-source models, two from the US and two from China. I’ll calculate the math and sketch the architectures by hand ✍️.
Week 1 — Gemma 4 (Google, USA): alternating global and local attention, kv-cache, sparse mixture of experts, per layer embedding.
👉 Week 2 — Qwen 3.6 (Alibaba, China): long-context scaling techniques and attention optimizations improving practical context length.
Week 3 — Nemotron 3 Super (NVIDIA, US): large-scale transformer systems, with emerging directions toward state-space alternatives like Mamba.
Week 4 — DeepSeek-V4 (DeepSeek, China): a 1.6T-parameter MoE with Compressed Sparse Attention for 1M-context and three test-time reasoning modes, advancing chain-of-thought at the open-source frontier, released just in-time for this special seminar series. 😉
February Seminars
2/5: Meta AI
Guest Expert: Yichen Wang (Meta)
2/12: Transformer
Guest Expert: Srijanie Dey, Tsavorite Scalable Intelligence
2/19: OpenClaw
Guest Expert: Val Andrei Fajardo, LlamaIndex Founding ML Engineer, Author of the Build a Multi-Agent System From Scratch
2/26: PPO➝DPO➝GRPO➝Rubrics
Guest Expert: Cameron R. Wolfe, Netflix, Author of Deep Learning Focus
January Seminars
1/29/2026: 9 AI Eval Formulas
1/22/2026: TPU
1/15/2026: How Small Models Learn Tool Use from AWS
1/8/2026: Manifold-Constrained Hyper Connection (mHC) from DeepSeek
1/8:2026: Introduction to Generative AI
About the Seminar Series
In 2026, I’ve made a personal commitment to teach regular live AI seminar by hand ✍️.
Foundation: intuition, math, and mental models of core AI concepts, for beginners.
Frontier: research papers, algorithm, and architectures used by frontier models, for advanced AI engineers and researchers.
Live attendance is free.
The recordings and Excel workbooks are available for members of the AI by Hand Academy.