dev.to

Earlier coverage

1. Shopify's public products.json stops paging at 25,000 products in live-store tests

   Build · September 29, 2026 · 1 publisher

2. A stale appearance stream lets a filled PDF form keep showing the value it replaced

   Build · September 27, 2026 · 1 publisher

3. Agent memory handoffs cut replaced context 98.65% by moving verbatim text out of the prompt

   Build · September 27, 2026 · 1 publisher

4. Resume-parsing guide: disclosure team owns the field allowlist, even if another team owns the template

   Build · September 27, 2026 · 1 publisher

5. Lookup-field improvement on a low-code platform broke a warehouse app untouched for eight months

   Build · September 27, 2026 · 1 publisher

6. A retry ID built from fixed state kept a trading bot's stop-loss repair failing for hours

   Build · September 27, 2026 · 1 publisher

7. One sanitizer strips invisible Unicode from every traveller's text before it reaches a prompt

   Build · September 27, 2026 · 1 publisher

8. A conditional SQL update should decide when a 10-minute inventory hold expires

   Build · September 27, 2026 · 1 publisher

9. Moving MCP secrets into a gateway leaves a hijacked agent holding only a per-run token

   Build · September 27, 2026 · 1 publisher

10. SvelteKit starter kits' boot-time migrations race once a second replica starts

    Build · September 27, 2026 · 1 publisher

11. A coding agent built a four-platform Kotlin messenger from a 33 KB plan cut into five steps

    Build · September 27, 2026 · 1 publisher

12. Agent-written code moves the first human review to the engineer who opens the pull request

    Build · September 27, 2026 · 1 publisher

13. Wall-clock last-write-wins overwrote an offline phone note while the phone's clock read correctly

    Build · September 26, 2026 · 1 publisher

14. One SQLite transaction keeps a retried Buy click from creating a second order

    Build · September 26, 2026 · 1 publisher

15. Reranker guide pairs every training example with a copy stripped of click statistics

    Build · September 26, 2026 · 1 publisher

16. Claude Code's subagent transcripts record which model and effort level each launch used

    Build · September 26, 2026 · 1 publisher

17. A release-gate scanner printed PASS for two months while scanning zero files

    Build · September 26, 2026 · 1 publisher

18. Dev.to fintech guide records checkout outcomes only after commit or rollback

    Build · September 26, 2026 · 1 publisher

19. Express upload guide moves the GPS-tag check to the re-encoded file it publishes

    Build · September 26, 2026 · 1 publisher

20. Per-tenant SPF, DKIM and DMARC checks give each publisher domain its own rollback point

    Build · September 26, 2026 · 1 publisher

21. Agent retries that feed errors back into the prompt make every attempt cost more than the last

    Build · September 26, 2026 · 1 publisher

22. A go/no-go rubric for MCP servers scores agent behavior separately from protocol tests

    Build · September 26, 2026 · 1 publisher

23. Coding-agent hard limits belong in scoped credentials and CI checks outside the system prompt

    Build · September 26, 2026 · 1 publisher

24. Your agent supplies its own confirmation: confirm=True is not a permission boundary

    Build · August 15, 2026 · 1 publisher

25. The outbox relay's missing integer: how one unsendable row stalls the whole pipeline

    Build · August 16, 2026 · 1 publisher

26. Andrew Pyle gives each coding agent its own git worktree cut from origin/main

    Build · September 26, 2026 · 1 publisher

27. Bounded label registries keep agent metrics from adding a time series per run

    Build · September 26, 2026 · 1 publisher

28. 157 plans, one real model: the expensive agent failures land before the first tool call

    Build · August 20, 2026 · 1 publisher

29. Checking a revision register keeps superseded documents out of RAG answers

    Build · September 26, 2026 · 1 publisher

30. A dev.to LLM retry pattern randomizes Retry-After waits between half and the full provider-requested time

    Build · September 26, 2026 · 1 publisher

31. Frontier models score at most 0.17 on a source-trust test that a two-line rule passes perfectly

    Build · September 26, 2026 · 1 publisher

32. A Kafka Streams auth cache kept a revoked API key working for four days

    Build · September 26, 2026 · 1 publisher

33. PostgreSQL's VACUUM strands sparse b-tree pages until a REINDEX rebuilds them

    Build · September 26, 2026 · 1 publisher

34. An explicit uncertainty gate removed 60% of one team's production agent incidents

    Build · August 29, 2026 · 1 publisher

35. Retrying a timed-out tool call can make an AI agent refund the same order twice

    Build · September 25, 2026 · 1 publisher

36. Summing planned pixels before rendering tells on-call whether a PDF preview retry will help

    Build · September 25, 2026 · 1 publisher

37. Adding a 'could not tell' result to checks exposed four new ways to misread them

    Build · September 25, 2026 · 1 publisher

38. A unique key written before the queue ack stops LLM retries from reordering moderation review

    Build · September 25, 2026 · 1 publisher

39. Stamping typeset PDFs cuts a 180,000-copy watermark run from 90 CPU-hours to 2

    Build · September 25, 2026 · 1 publisher

40. A 4-bit 7B model that fits in 4 GB still runs out of memory near 30,000 tokens

    Build · September 25, 2026 · 1 publisher