~/talk

Agentic AI and the Next Decade of Quality Engineering

on YouTube 37m
Talkat TestMu Conference 2026 Β· 2026-08-19..08-21Β· [ai-assisted-development, engineering-leadership]

speaker

name:𝐌𝐚π₯π₯𝐒𝐀𝐚 π…πžπ«π§πšπ§ππžπ¬
links:none on file

resources

video:https://youtu.be/4a9yHZMF1nM
slides:none on file

summary

Argues agentic AI moves repetitive QE work to self-generating test suites that design, write and execute tests, convert legacy suites to Playwright/Selenium, and even run reviewer agents that regenerate code when reviews fail; demonstrates use cases for test design, data generation and risk-based selection, warns of token-cost traps with a Concorde analogy, cites current 30–50% progression automation and projects 30–40% lower cost and self-healing execution if humans enforce guardrails and PII controls.

AI-written summary Β· about

chapters (40)

  1. 0:00Agentic AI and the Next Decade of Quality Engineering
  2. 2:11Speaker Intro - Mallika Fernandes, Managing Director – Quality Engineering at Accenture
  3. 3:49What's Here Today, What the Risks Are, What the Decade Holds
  4. 4:53The Three Masons: Cutting Stone, Building a Wall, Building a Cathedral
  5. 5:57What Every Tester Is Actually Contributing To
  6. 6:27The Evolution: Manual Testing, WinRunner and Record-Replay
  7. 6:582010 and Agile: Shift Left, In-Sprint Testing, Progression Testing
  8. 7:30The Struggle to Automate Progression, Not Just Regression
  9. 8:00From Unattended Automation to Quality Engineering as a Mindset
  10. 9:00Classical AI as an Assistant, Then Generative, Now Autonomous
  11. 10:30Where QE Is Heading: Self-Generating, Self-Executing Suites
  12. 11:00Why the Mindset Has to Change Alongside the Technology
  13. 11:32Why Music Bands Became Less Common - and What It Means
  14. 13:06Automation Let One Artist Do What a Band Used to Need
  15. 13:37The Same Shift in Software: Mechanical Work Moves to AI
  16. 14:07What That Leaves Us: Planning, Design, Strategy, Innovation
  17. 15:10"Talk Is Cheap, Show Me the Code" Becomes "Code Is Cheap, Show Me the Outcome"
  18. 16:12Agent Use Cases: Test Design, Scripts, Data, Risk-Based Selection
  19. 16:42Modernising Legacy Test Suites Into a New Framework
  20. 17:13Why Context Building and Prompt Engineering Decide Accuracy
  21. 17:43Reviewer Agents That Regenerate Code When the Review Fails
  22. 18:15How to Adopt AI: As a Tool in Your Kit
  23. 19:17Treat It Like an Enthusiastic Intern Who Works Twelve Hours
  24. 20:20Will AI Take Our Jobs? Pivot to Higher-Order Work
  25. 21:22Risks and Edge Cases: The Concorde Story
  26. 22:27An Engineering Marvel That Failed on Cost Economics
  27. 23:30The Parallel: Token Economics and Client ROI
  28. 24:32Five Stages of AI Usage - and Where It Gets Uncomfortable
  29. 25:36Stage Two: AI as a Pair, the Level She Recommends
  30. 26:06Stage Three: Automation Bias, Accepting Output Without Question
  31. 26:37South Korea's Law Mandating Human Involvement
  32. 27:09Stage Four: Skill Atrophy - The Google Maps Confession
  33. 27:40Stage Five: Cognitive Surrender
  34. 28:12The Next Decade: Dark Factories and the Human Auditor
  35. 29:13AI-Driven QE: 30-40% Lower Cost, True Shift-Left Intelligence
  36. 30:15From Maintenance Burden to Self-Healing at Execution
  37. 30:47From Quality to Trust Engineering
  38. 31:48Q&A: What to Stop Doing Manually and What Skill to Build First
  39. 33:51Q&A: The Failure Mode Teams Walk Into Most Often
  40. 34:51Q&A: Agents Colluding, Guardrails and PII Protection

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 β€” the status bar button does the same
?
this help
esc
close this, then clear every filter β€” 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.