~/talk

The paradox of perfect process

links:none on file
slides:none on file

A global ML recommendation engine plus AI image recognition ships from a separate POC into the main app and reaches 20 million daily users during a 10-month global rollout, showing 35% conversion uplift on surface metrics while the purchase rate stays flat. Root causes include payment-step data that captures only a fraction of purchases, poor recency weighting, and a feedback loop where confusing recommendations amplify irrelevant clicks. Fixes: treat low-history users as new, add market-popular lists, human curation and stricter signals, and prioritize purchases over clicks.

AI-written summary · about

  1. 0:00Introduction and context
  2. 3:12From payments to shopping
  3. 6:43Why the delivery succeeded
  4. 11:43The hidden failure
  5. 14:13Root causes analysis
  6. 20:14Customer journey mismatch
  7. 22:43Testing beyond happy paths
  8. 26:14Fixing the machine
  9. 29:07Key takeaways
  10. 30:51Conclusion and Q&A
~/search
type to search 6,518 conferences, 143,918 speakers and 192,133 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
← →
on a talk: 5 seconds back / forward
0–9
on a talk: jump to 0%–90% of the recording
?
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.