Daniel Cummings

I’m an AI/ML engineer and researcher with a background in computer engineering. My recent work uses AI to help design the chips that AI runs on, with agentic systems, LLMs and ML optimization across spec to GDS, design verification and post-silicon. It’s a little circular, and I enjoy that.

Chip Field Guide is a personal project. It isn’t affiliated with or endorsed by my employer, and the views here are my own.

Chip Field Guide is a set of free, interactive guides to how chips work. 4 guides run from the transistor to the datacenter: Transistors, Design Flow, Architectures and Systems. Every chapter has a small simulation you can play with.

• Three depths. Read any page as a Beginner (no background needed), a Novice (new to some of the topics) or an Expert (knows the basics). The switch at the top sets the whole site, and each section has its own switch.
• Hands-on. Flip a transistor on and watch the channel form, read an SRAM cell until it flips, synthesize logic, route nets, step a systolic array through a matrix multiply, plan how a model splits across a thousand accelerators, and budget a liquid-cooled rack, right in the browser.
• Open sources. Every factual claim links to a source you can read for free: open-source tool docs, open papers, course notes and public references. Examples use open-source tools such as Yosys and OpenROAD.
• Extras. 7 industry pages, a tool landscape, AI in EDA, what happens after tapeout and a glossary.
• No accounts. Your detail level, quiz answers and progress stay in your browser.

If Chip Field Guide helped with a paper, a course or a talk, you can cite it with the entry below. To cite one chapter, swap in that page’s title and address, and add the date you read it. The text and figures are shared under CC BY-NC 4.0.

Spotted a mistake, a broken link or a topic that deserves a chapter? I’d like to hear about it. The quickest way to reach me is on LinkedIn. Or email daniel@chipcodex.com.