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-hassan-agentic-engineering/references/

Source outline

Source: Ahmed E. Hassan, The Tool Got Smarter. Did You?, AI Native DevCon New York 2026. Recording. All anchors refer to original numbered cues in the transcript. Short quotations below preserve the supplied transcription, including spoken wording.

TopicSource anchorExact short quoteApplication and scope
Engineering system00:00:27–00:01:51, cues 8–32“you really need an engineering system.”Improved tools still require an engineered process.
Immersion and local limits00:02:21–00:04:47, cues 41–82“benchmarks don't represent you.”Track task-specific limits and learn through bounded attempts.
Accountability00:04:49–00:07:31, cues 83–130“You need to review the workflows.”Middle-loop engineering concerns process ownership.
Compounding corrections00:07:33–00:08:30, cues 131–145“that correction will build up.”Human interventions should improve future attempts.
Failure learning00:09:15–00:09:58, cues 157–167“every rescue should improve the next attempt.”Guidance or verification tools can capture lessons.
Verification river00:10:41–00:12:23, cues 180–207“don't be the hands and the eyes of the agent,”Invest in appropriate autonomous verification.
Retest workarounds00:12:41–00:14:35, cues 212–247“is this a stopgap solution”Retest real tasks and distinguish temporary hacks from durable needs.
Responsibility design00:14:43–00:16:35, cues 250–280“Define what AI may execute, what it may decide,”Specify authority and escalation.
Mixed technologies00:16:40–00:18:36, cues 282–317“the total cost per result.”Combine suitable tools and compare cost and errors.
Restraint00:18:39–00:22:11, cues 318–379“a huge difference between a demo and shipping software.”Capability does not remove production work or human limits.
Trust and knowledge Q&A00:23:58–00:26:03, cues 403–436“it's a combination of trust and knowledge.”The speaker allows tighter engagement for important parts.
Education Q&A00:26:14–00:29:58, cues 438–495“We added a course for reading code.”Educational changes are reported experiments and discussion.

Source limitations

  • The principles are recommendations and opinions; the transcript does not establish universal model trajectories, productivity gains, or an error-free process.
  • Historical analogies and the informal ISO explanation are not standards guidance; do not use them to assert compliance requirements.
  • The cited research about newer models lacks enough detail here for independent verification or precise generalization.
  • Q&A includes audience opinions and debated education approaches. Do not attribute all audience claims to Hassan or treat junior development as solved.
  • Promotional interlude cues 146–156 are omitted from the bundled transcript; the remaining cue numbers retain their source positions.
  • This is a spoken transcript; unseen slides, linked repositories, and third-party research were not independently inspected. No outcome beyond the described talk is established.

skills

talk-hassan-agentic-engineering

tile.json