Software Engineering 52
- specs are not agent memory
- why your ai agent fails: it is a harness problem, not a prompt problem
- DBOS and the shrinking cost of durable workflows
- Herdr is the throughput layer for coding agents
- NATS is the messaging backbone I reach for before Kafka
- better tools can make ai code review worse
- the ai learning shortcut may become apprenticeship debt
- agentic autofix needs evidence, not applause
- ai-era interviews should test evidence, not leetcode muscle memory
- choosing the right agent surface: Codex, Claude Code, Copilot, and when not to use an agent
- review and verification: making agent code mergeable
- subagents and specialized agents: parallelism without chaos
- instructions as code: AGENTS.md, CLAUDE.md, and copilot-instructions.md
- the senior engineer's agent loop: context, plan, patch, proof
- how senior engineers should code with agents without becoming reviewers of chaos
- coding benchmarks are becoming production dependencies
- agent kits are turning developer docs into executable guardrails
- incident response is becoming an agent review workflow
- copilot browser tools make the frontend reviewable
- the best books for the agent era are still systems books
- frontier models are becoming cloud procurement
- Agentic incident response is where autonomy meets the pager
- agents need shared memory, not another context window
- AI code governance is the new code review bottleneck
- VEX turns container scanning into queue discipline
- AI credits are the new lines of code metric
- AGENTS.md is becoming the new code review contract
- agent evaluation is becoming the new test pyramid
- GitHub Actions is becoming the agent runtime
- prompts are becoming CI/CD configuration
- coding agents made repositories the security boundary
- agent canvases are the end of chat-only coding tools
- copilot automations turn agents into scheduled infrastructure
- AI code review is now a cloud workload
- postgres is all you need for durable execution
- open-source coding agents need maintainers, not just models
- ai productivity gains are real, but smaller in production
- cognitive debt is the ai code smell nobody wants to measure
- Watch out - that recruiter profile might be a scam. Check here first.
- How to safely inspect a repository before running anything on your machine
- replit chased the ai hype and left a simple coding gap
- why the 'hand coding' backlash is really about agency, not nostalgia
- measuring ai-assisted velocity without lying to yourself
- ai agent fluency is the new staff+ engineering skill
- AI Tools Have Shorter Half-Lives Than the Workflows They Automate
- Tactical Debt Is the Silent Killer of Engineering Velocity
- Tactical Debt Silently Destroys Engineering Velocity
- Hyped and Overhyped Programming Languages in 2026
- Hyped and Overhyped Programming Languages in 2026
- your second brain should not be a folder full of markdown
- github failed at the only thing they should do: git
- why agi is not possible with the current llms and transformers