01 / Writing
Writing
Everything, in one place, newest first. The tag marks where a piece first appeared; the canonical copy lives here. Imports keep their original text — typos included. It's an archive, not a rewrite.
2026
- 2026-10 Agents need a factory floor, not just a model linkedin
- 2026-10 Semantic Model Portability Is More Than a Niche Concern linkedin
- 2026-10 Replace constraints, not everything linkedin
- 2026-10 MCP Is an Unstable Contract Until You Test Your Client linkedin
- 2026-10 The quiet shift to right-sized models in security pipelines linkedin
- 2026-09 Docker cloud sandboxes keep agents running after you close the laptop linkedin
- 2026-09 AI proposes, engine proves: a handoff standard for chip design linkedin
- 2026-09 Pinecone BYOC: managed vector data in your cloud linkedin
- 2026-09 Enterprise copilots meet all three conditions for silent exfiltration linkedin
- 2026-09 Sweeping codebase changes feel safe until they are not linkedin
- 2026-09 Access gateways and serving gateways are not the same thing linkedin
- 2026-09 What Graph RAG Changes About Code Migration Retrieval linkedin
- 2026-09 Westpac's Adapt is a controlled migration pattern, not a destination linkedin
- 2026-09 The Left Boundary in Shift-Left Has Moved linkedin
- 2026-09 The MiniMax M3 optimization lesson: chasing the moving bottleneck linkedin
- 2026-09 Supervised agents beat autonomy for legacy modernization linkedin
- 2026-09 Shuffle-sharding: probabilistic isolation without dedicated hardware linkedin
- 2026-09 Human-in-the-loop AI needs real override authority linkedin
- 2026-09 AI agents won't replace the system of record—they'll wrap it linkedin
- 2026-09 The coming provider abstraction layer in enterprise AI linkedin
- 2026-09 Modernization is subtraction, not addition linkedin
- 2026-09 IT governance was built for recommendations, not actions linkedin
- 2026-09 Elastic partitioning redefines Kafka partitions as movable units linkedin
- 2026-08 The availability risks that never go away linkedin
- 2026-08 Financial services agents will move from pilot to standard tooling linkedin
- 2026-08 Why I'd instrument before building an internal AI cluster linkedin
- 2026-08 The coordination tax in a polyrepo estate linkedin
- 2026-08 Open-weight models are moving into production workloads linkedin
- 2026-08 Slack Code moves AI coding out of the terminal into group chat linkedin
- 2026-08 Boring Infrastructure Sleeps Better linkedin
- 2026-08 Enterprise AI’s Quiet Shift from Experiments to Unit Economics linkedin
- 2026-08 An LLM wiki is not a self-maintaining knowledge base linkedin
- 2026-08 MCP session handles are capability tokens, not pointers linkedin
- 2026-08 Why Regulated Enterprises Will Mix Open and Closed Models linkedin
- 2026-08 Multi-cloud AI observability needs a common transponder linkedin
- 2026-08 AI-assisted PostgreSQL migrations need a human review loop linkedin
- 2026-08 The Real Bottleneck in Data Platforms Is Neglected Orchestration linkedin
- 2026-08 Why I’d Bet on the Agent Access Model for Production linkedin
- 2026-08 Memory is the bottleneck nobody budgets for linkedin
- 2026-08 The enterprise AI strategy that outlasts any single model linkedin
- 2026-08 Agents Are Easy to Build, Hard to Control linkedin
- 2026-08 Overprivileged AI Agents Repeat Old IAM Mistakes linkedin
- 2026-08 Stateful Execution Is the Next Bottleneck for Enterprise AI Agents linkedin
- 2026-07 Observability Is a Safety Net, Not a Silver Bullet, for Vibe Coding linkedin
- 2026-07 Routine coding tasks don't need expensive reasoning models linkedin
- 2026-07 Model AI agents as platform citizens now, not special cases later linkedin
- 2026-07 Set Runtime Constraints on AI Agents Before They Hit Production linkedin
- 2026-07 The Meaning Wall: Why AI Agents Need a Knowledge Layer linkedin
- 2026-07 Centralized AI gateways: necessary but not sufficient linkedin
- 2026-07 Try it at home first site
- 2026-07 Why this site is boring on purpose site