The strongest ai-coding story this month isn't in software, it's in silicon. china's empyrean cut a…
the strongest ai-coding story this month isn't in software, it's in silicon. china's empyrean cut a chip design task from four weeks to one with agentic ai, and samsung trimmed some soc verification from a month to two days using claude.
the eda flow is where agents stop being a demo.
Context
Two separate single-task speedups are reported. XenoSpectrum, relaying a South China Morning Post report of September 13, 2026, says Empyrean's chairman said an AI agent cut a circuit layout task from four weeks to one, and Wccftech says as much as 75% for some tasks. ChosunBiz reported on August 12, 2026 that Samsung's System LSI division used Claude and finished a task expected to take more than a month in two days, for verification of a customized SoC's data interconnect, which Samsung assessed as about 15 times faster. TechRadar's headline on August 25, 2026 adds that the AI still makes some worryingly big mistakes.
These are different tasks at different firms, layout and verification, so they cannot be compared. The Empyrean figure is one example stated by the company chairman with no stated benchmark, baseline or verification, and the original SCMP report was not read. The Samsung figure is one project reported through the press. The Samsung item dates from August 12, so this month fits only Empyrean. The strongest story this month is the author's take.
Watch next
- The SCMP original, Empyrean's own release, and Samsung's confirmation of the 15x figure and error rates.
Sources
- XenoSpectrum: Empyrean AI layoutxenospectrum.com
- Wccftech: Empyrean slashes circuit design timewccftech.com
- ChosunBiz: Samsung and Claude (August 12, 2026)biz.chosun.com
- TechRadar: Samsung and Claude Codetechradar.com
Provenance
The note above is reproduced unedited from the original post, first published on Threads on 21 September 2026 at 21:16 IST. Sources are the papers and datasets the note draws on.
View the original post ↗Embed this note
More notes
The air is now being asked to keep its own ledger
the air is now being asked to keep its own ledger: ecmwf’s aifs compo becomes the first ai model to forecast atmospheric composition globally every three hours, cleanair simulates 365 days of pm2.5 over china in ten seconds, and a unified framework maps six pollutants at one kilometer across the whole country. the air now files its own composition report.
read the note →The current is now being asked to draw its own map
the current is now being asked to draw its own map: china’s langya 2.0 predicts six ocean phenomena including internal waves and mesoscale eddies, a deep net called wenhai resolves eddies globally with air sea flux formulas built in, and scripps infers surface currents from the way temperature patterns deform in satellite images. the ocean now files its own circulation report.
read the note →The soil is now being asked to report its own carbon
the soil is now being asked to report its own carbon: a nix color sensor paired with generative data augmentation predicts soil organic carbon without a lab, random forest drives 74 percent of soil health mapping studies, and sentinel 2 tracks five year carbon change across france and italy from 922 samples. the dirt now files its own carbon account.
read the note →