~/talk

Scaling Quality in a Decision Intelligence Platform: The Agentic QA Playbook

on YouTube 31m
Talkat TestMu Conference 2026 · 2026-08-19..08-21· [ai-assisted-development, testing]

speaker

name:Hammad Ahmed
links:none on file

resources

video:https://youtu.be/3K8TEut31fU
slides:none on file

summary

Aily Labs rebuilds QA into an always-on agent network that ingests 10,000 test cases, Confluence and Jira docs, and uses Playwright, Appium, Behave, Datadog and Langfuse to ground behavior; separate agents perform exploratory testing, resiliency checks, PR review, RCA, criticism and bug fixing, and they swarm and autonomously link so RCA can trigger a bug fix and create a regression test; a quality pulse tracks accuracy, defect leakage and tenant health; smaller specialist models cut evaluator latency and cost; Slack and GitHub integration delivers daily proactive briefings and on‑demand runs.

AI-written summary · about

chapters (35)

  1. 0:00Scaling Quality in a Decision Intelligence Platform
  2. 1:40Speaker Intro - Hammad Ahmed, Head of AI QA Innovation, Aily Labs
  3. 2:42The Breaking Point: What Aily Builds and Why Testing Broke
  4. 3:12From Manual Excel Analysis to AI-Driven Decision Intelligence
  5. 4:42The CEO's Challenge: Scaling Fast With Fortune 500 Clients
  6. 5:13An Application Whose Behaviour Isn't Predictable
  7. 5:45No Room for Error Across Verticals, Platforms and Devices
  8. 6:45The Idea: Build a Quality Super Agent Internally
  9. 7:15Starting With an Exploratory Testing Agent
  10. 7:46Adding Resiliency, Test Creation, PR Review and RCA Agents
  11. 8:17The Critic Agent and the Bug-Fixing Agent
  12. 8:49Swarming the Agents and Letting Them Talk to Each Other
  13. 9:21Reusing Conventional Tools: Playwright, Appium, Behave, Datadog, Langfuse
  14. 10:22Grounding the Agents With Confluence, Jira and Product Specs
  15. 10:52Feeding 10,000 Existing Test Cases as Product Documentation
  16. 11:53Documenting Business Rules for Industry-Specific Knowledge
  17. 12:55The Context Got Bigger and Hallucination Increased
  18. 13:26Adopting Skill Gating to Access Only the Relevant Context
  19. 14:29Small Silos of Knowledge: Financial, R&D, Manufacturing, Supply
  20. 15:31Building a Quality Pulse: Accuracy, Defect Leakage, Tenant Health
  21. 16:32Signals From the Exploratory, PR Review and RCA Agents
  22. 17:02Adoption: Embedding the Agents Into Slack and GitHub
  23. 17:32Moving From Reactive AI to Proactive Daily Briefings
  24. 18:32Prioritisation for Leads in a Fast-Scaling Company
  25. 19:34On-Demand Runs for Anyone in the Company
  26. 20:04A Super QA Engineer Live Inside the Workflow
  27. 20:36What's Next: Agents Talking to Each Other Autonomously
  28. 21:38Connecting RCA to Bug Fixing to Test Creation
  29. 22:39Takeaway One: Give Quality a Pulse
  30. 23:09Takeaway Two: Start Building and Orchestrating Agents
  31. 24:09Takeaway Three: Adopt AI Into the Workflow, Not as a Shortcut
  32. 25:40Q&A: Optimising LLM Evaluator Latency and Cost With Smaller Models
  33. 27:16Q&A: Catching a Triage or RCA Agent That Gets It Wrong
  34. 28:48Q&A: Are We Creating an Accountability Gap?
  35. 29:19Q&A: Put the Guardrails In Before You Expose AI to Any Repository

related talks

~/search
type to search 6,249 conferences, 138,987 speakers and 185,172 talks
~/help
cmdctrl k
search every conference, speaker and series
/
focus the filter search
h
home
c
conferences
s
speakers
e
series
t
light / dark — the status bar button does the same
?
this help
esc
close this, then clear every filter — or just the search, while you are typing in it

Ignored while you are typing. Pages without a filter rail have no search to focus.