From the desk of Apex research + engineering
Two-minute read; links to the underlying papers, model cards and case studies inline.
01 · Research progress
Full RTL → GDS flow on ASAP7 predictive PDK now ships with WNS ≤ 0.4% gap vs commercial EDA. Methods paper at /research.
02 · Customer wins
Fabless customer co-authored a comparison memo on the ASAP7 result. Quote: 'We expected to see Aether as a sidecar; six weeks in, it's the primary path on two design families.'
03 · Model-card diffs
A subtle convergence bug in the FEA solver was caught by a customer and fixed within a week. Postmortem on /status. Refusal corpus unchanged.
04 · An open question
What's the longest-standing bug in your customer's incumbent stack? We can't see your customer; you can. Reply.
05 · What we read
- (1) On predictive PDK quality vs foundry PDK — Kowalski.
- (2) Customer comparison memo — fabless customer.
- (3) FEA convergence bug postmortem — engineering.