~/talk

Agentic AI and Code Reviews: Patterns and Case Studies

speakerdeck:mlgill
slides:none on file

AI-generated code rises 19% to 57% in three quarters, and top teams now move 9x faster than the median. The presentation lays out concrete patterns: small incremental changes, design for verification and spec-driven development, context engineering, deterministic checks, and adversarial and multi-agent review. DryRun Security fixes a missed multi-tenant isolation by adding product-specific context and spec checks; Cloudflare shifts reviews into CI with an orchestrator to dedupe noisy feedback; Block pilots local AI reviews then scales to merge-ready automated fixes. These practices remove review as the bottleneck.

AI-written summary · about

  1. 0:00Introduction by Gene Kim
  2. 1:14AI code surge and 9X teams
  3. 3:13Why code review still matters
  4. 6:48Patterns for agentic code review
  5. 14:27DryRun Security case study
  6. 17:56Cloudflare case study
  7. 21:14Block case study
  8. 23:45Takeaways and call for help
~/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.