Another great day of learning and building here in Tokyo. There’s an insane number of things happening in tech this week, and I’m barely keeping up.
[blog] Gemini Decisions API: Fast, calibrated multimodal decisions. We’ve got a decision model. Send in all sorts of input along with independent questions, and get back probabilities and confidence scores.
[blog] Beyond Writing Code: Taste, Skills, and Tech Leads. We’re all tech leads now, but where does the human work come in? Daniela has a good take.
[blog] The future of Software Engineering is boring. I get what Alex is saying. At the same time, I’m enjoying software engineering now more than ever. But we’ll need to actively adjust to ensure we keep learning, and don’t become meat proxies.
[blog] The components of agent memory. When using offline compute (“dreaming”) to reflect on session transcripts, AI agents can consolidate experiences into structured notes and catch errors.
[blog] We gave agents better instructions. Cool idea from the team at Mintlify. This new CLI command gives guidance to agent that greatly reduces hallucinations.
[blog] New survey finds enterprises trapped between legacy software and AI modernization. Companies are held back by previous investments. That’s nothing new, but the urgency is higher now. Interesting data!
[article] Anthropic and OpenAI are coming for the development workflow. I hear Matt’s point here. You probably don’t want to pick fundamental infrastructure (e.g. source code storage) based on which model you like this month.
[blog] How Salesforce Scaled Engineering AI Adoption Without Stifling Innovation. I like these types of experience reports coming from big shops, not just small startups with no preexisting tech commitments.
[blog] Don’t just prompt AI — architect it: Introducing the Professional Agentic Architect Certification. Test your agent skills with this comprehensive certification exam. Looking at the topics covered, it’s a legit test of knowledge.
[article] AWS gives developers a free local stand-in for its cloud, built from scratch by AI agents. An agent-built emulator for AWS? Pretty cool work. Some docs here.
[article] Redesigning the Dashboard: Cultivating Trust Over Gamification in AI Adoption. Our Google developer experience team learned some things about token tracking dashboards.
Want to get this update sent to you every day? Subscribe to my RSS feed or subscribe via email below:
Leave a Reply