CtrlK
BlogDocsLog inGet started
Tessl Logo
PluginPublic

ainativedev/aidevcon-2026-nyc

AI Native DevCon 2026 New York — practical agent skills grounded in conference talks

Quality

86%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Low

Low-risk findings worth noting

Overview
Quality
Evals
Security
Files

outline.mdskills/talk-gavrilescu-factory-architecture/references/

Source outline

Use transcript.md for exact evidence. Promotion cues 195–205 are omitted.

TimeCuesTopic
00:00–02:131–29Tool motivation and personal throughput measure
02:15–05:3730–72Unfamiliar production outage and discovered credential
05:41–07:5473–103Lost understanding and accumulated duplication
07:57–09:45104–128Cleanup, recurrence and PR-local review limits
09:48–10:59129–143Stale diagrams and desired architecture/product review
11:01–13:46144–178Failed iterations and deterministic scanner approach
14:14–14:55185–194Intent-to-architecture review loop
15:38–17:28206–232C4 demo, curated flows and stable IDs
17:33–20:03233–272Greenfield demo and scanner discussion
20:08–22:12273–303Keycloak raw scan and curated brownfield baseline
22:18–25:35305–349Diffs, static exports and task integration
27:21–30:53369–425Layout, sequence-diagram and cross-language limits

Short quotations

  • “I had no idea where to start” — cue 44, 03:24.
  • “every PR is fine on its own.” — excerpt, cue 123, 09:17.
  • “I cannot invent things” — excerpt, cue 226, 17:00.

Limits

Reported recovery times, the $8 demo spend and commit counts are not general benchmarks. C4 viewer compatibility is suggested but explicitly untested. The demo's enriched flows involve human or agent curation; they are not wholly inferred by deterministic scanning. Sequence diagrams are described as not yet available. Q&A about Python coverage is partially garbled; avoid a blanket compatibility claim.

skills

talk-gavrilescu-factory-architecture

tile.json