Thoughts on AI, engineering, and building
Practical writing on AI agents, automation, and the engineering behind them, for the people who run the business.
Security & Governanceexplainer
A plain-English map of the data an AI request sends to a vendor, how long it can be kept, and the four questions to settle before sensitive data goes in.
· 4 min read
Hardware & Infrastructurehardware-breakdown
AI clusters use two networks: one inside the rack and one between racks. Specs from Nvidia's Rubin platform and the Ultra Ethernet standard.
· 3 min read
Harnesses & Toolingdeep-dive
Models now plan spreadsheet work and read business documents. Why that raises, not lowers, the value of clean pipelines and anomaly detection.
· 4 min read
Models & Benchmarksexplainer
Open-weight models let you run and modify the model yourself; closed models are rented by API. A plain-English comparison and a decision guide for businesses.
· 3 min read
Harnesses & Toolingbrief
Embeddings turn text into numbers so software can search by meaning. What they are, where they are used, and what to ask a vendor about them.
· 2 min read
Economics & Pricingbrief
Challenger's January report attributes 7,624 U.S. job cuts to AI, 7% of the total. What the figure shows, and what its authors say it does not.
· 1 min read
Agentsnews-analysis
OpenAI's GPT-5.3-Codex scored 64.7% on OSWorld-Verified, up from 38.2%. What the number means for browser and desktop agents, and what it leaves out.
· 3 min read
Security & Governancechecklist
A practical 12-question list for evaluating an AI vendor: data handling, testing, oversight, agent permissions and exit terms.
· 2 min read
Models & Benchmarksexplainer
A plain-English guide to AI evals: test sets, graders, repeated trials and why a 75 percent success rate fails more often than it sounds.
· 4 min read
Models & Benchmarksbrief
AI benchmarks are saturating and scores depend on test setup. Three published examples, and how to read a model announcement without being misled.
· 2 min read
Enterprise Adoptiondeep-dive
What published surveys show about AI use among contractors and construction firms, where it pays off first, and where the evidence is thin.
· 4 min read
Agentsexplainer
A plain-English guide to designing human oversight into AI workflows, with a checklist of what reviewers need and this week's desktop agent launches.
· 3 min read