From “CEO said a thing” journalism and sovereign payments to product teams in code-gen land, Deming vs. OKRs, sex recessions, Neanderthals, fonts, and hype cycles - a tour through power, productivity, and what actually holds systems together.
Posts in "tech"
Your Boss Doesn’t Know What to Do With AI
Enterprise AI Has a Product-Market Fit Problem. Enterprise AI isn’t stalled because the models are weak. It’s stalled because we haven’t discovered product-market fit inside the enterprise yet.
You don’t find real AI value by theorizing in workshops. You find it by running experiments for months inside your actual systems - against real data - in a governed environment.
That requires a platform.
Without one, AI pilots turn into disconnected experiments, shadow infrastructure, and compliance risk. With one, experimentation compounds into institutional learning.
In this video, I break down:
- Why enterprise AI is still in discovery mode
- Why experimentation must be long-running, not one-off
- How governance enables innovation instead of blocking it
- Why a secure platform foundation is the baseline for AI ROI
If you’re thinking about AI strategy, platform engineering, or how to make AI experimentation safe and scalable, this is where to start.
Featured:
There's a lot of business logic in Java, decades worth...
We have invested a lot in domain models, some of which are even very good. And, to be able to leverage that as we move to the new world is really, really important." Rod Johnson.
Source: “GenAI Grows Up: Building Production-Ready Agents on the JVM," Rod Johnson, GOTO, October 1st, 2025.
That reminds me of this gem:
Tanzu Catsup: The Risk of Relying on AI for Platform Engineering
Internal development teams and executives are increasingly looking at AI to automate the creation of internal developer platforms. However, cobbling together open-source components with AI is a far cry from building a scalable, secure, and “enterprise-grade” environment. In this conversation, we explore why betting your internal infrastructure on AI-generated platforms is a high-stakes risk and why human-led architecture still dictates the long-term success of a platform.
There’s also an excerpt.
Related to your interests, Thursday
Europe doubles down on sovereign cloud, Anthropic shakes markets, and AI seeps into meetings, legislation, and IT services. Plus: Java for AI, portable agent “skills,” ERP disdain, and cultural scale reality checks.
Enterprise AI needs new apps, enterprise AI doesn't need new platforms
Enterprises should focus on resilience and learning in AI projects by experimenting with applications rather than building new infrastructure.
The weird bastion stuff in D&D 5e 2024 makes a lot more sense after reading Dungeon Crawler Carl.
Go cheap on innovation. Otherwise you just buy PowerPoints. Jana Werner on using tight timelines and small teams to force real progress.
Two theories of the AI investor spook-out
Concerns are growing that AI-generated applications could disrupt existing software companies, leading to significant drops in market value for tech stocks, particularly following major announcements from firms like Anthropic and Amazon.
Relative to your interests, Friday
Enterprise AI reality checks everywhere: no more free lunches, Copilot productivity gaps, agent sprawl, usage control, and OSS maintainers drowning in AI slop. Also: Kubernetes + Greenplum, MCP auth that finally works, EA’s quiet resurgence, AI prompts-as-infra, propaganda via images, and yes, a Chicken McNugget futures market.