vidtheque.

read-only demo

Videos VMemhtlsoNk

Agents Don't Do Standups: Building the Post-Engineer Engineering Org — Mike Spitz, PFF

index_state ready data_status ok

AI Engineer· published 2026-05-15· 0:17:49· en-US· indexed 2026-08-10 19:51

Open on YouTube

Scene timeline

  1. Shot 0, 0:00 to 0:05, 1 of 1 keyframes kept
  2. Shot 1, 0:05 to 0:09, 1 of 1 keyframes kept
  3. Shot 2, 0:09 to 0:14, 1 of 1 keyframes kept
  4. Shot 3, 0:14 to 0:46, 1 of 1 keyframes kept
  5. Shot 4, 0:46 to 1:12, 1 of 1 keyframes kept
  6. Shot 5, 1:12 to 1:37, 1 of 1 keyframes kept
  7. Shot 6, 1:37 to 2:03, 1 of 1 keyframes kept
  8. Shot 7, 2:03 to 2:29, 0 of 1 keyframes kept
  9. Shot 8, 2:29 to 2:57, 1 of 1 keyframes kept
  10. Shot 9, 2:57 to 3:21, 1 of 1 keyframes kept
  11. Shot 10, 3:21 to 3:37, 0 of 1 keyframes kept
  12. Shot 11, 3:37 to 4:02, 1 of 1 keyframes kept
  13. Shot 12, 4:02 to 4:28, 0 of 1 keyframes kept
  14. Shot 13, 4:28 to 4:54, 1 of 1 keyframes kept
  15. Shot 14, 4:54 to 5:27, 0 of 1 keyframes kept
  16. Shot 15, 5:27 to 6:00, 1 of 1 keyframes kept
  17. Shot 16, 6:00 to 6:33, 0 of 1 keyframes kept
  18. Shot 17, 6:33 to 6:54, 0 of 1 keyframes kept
  19. Shot 18, 6:54 to 7:22, 0 of 1 keyframes kept
  20. Shot 19, 7:22 to 7:51, 0 of 1 keyframes kept
  21. Shot 20, 7:51 to 8:19, 0 of 1 keyframes kept
  22. Shot 21, 8:19 to 8:47, 0 of 1 keyframes kept
  23. Shot 22, 8:47 to 9:15, 1 of 1 keyframes kept
  24. Shot 23, 9:15 to 9:34, 1 of 1 keyframes kept
  25. Shot 24, 9:34 to 10:08, 1 of 1 keyframes kept
  26. Shot 25, 10:08 to 10:36, 1 of 1 keyframes kept
  27. Shot 26, 10:36 to 11:05, 0 of 1 keyframes kept
  28. Shot 27, 11:05 to 11:34, 1 of 1 keyframes kept
  29. Shot 28, 11:34 to 12:03, 0 of 1 keyframes kept
  30. Shot 29, 12:03 to 12:34, 0 of 1 keyframes kept
  31. Shot 30, 12:34 to 13:05, 0 of 1 keyframes kept
  32. Shot 31, 13:05 to 13:37, 0 of 1 keyframes kept
  33. Shot 32, 13:37 to 14:08, 1 of 1 keyframes kept
  34. Shot 33, 14:08 to 14:47, 0 of 1 keyframes kept
  35. Shot 34, 14:47 to 15:17, 0 of 1 keyframes kept
  36. Shot 35, 15:17 to 15:46, 0 of 1 keyframes kept
  37. Shot 36, 15:46 to 16:16, 0 of 1 keyframes kept
  38. Shot 37, 16:16 to 16:23, 1 of 1 keyframes kept
  39. Shot 38, 16:23 to 16:49, 0 of 1 keyframes kept
  40. Shot 39, 16:49 to 17:15, 0 of 1 keyframes kept
  41. Shot 40, 17:15 to 17:34, 0 of 1 keyframes kept
  42. Shot 41, 17:34 to 17:49, 1 of 1 keyframes kept

42 shot(s).

keyframes kept every frame deduplicated

What was stored

cues
187
whisperx 187
chunks
30
from 187 cues
keyframes
20
kept of 42 captured
frames with text
20
406 lines read
chapters
13
from the source metadata
keyframe bytes
4.2 MB
word timings on 187 cues

Provenance

Each pipeline stage, its state and the model that produced it
stage state model started took
fetch done — 2026-08-10 14:23 2m 05s
stt done — 2026-08-10 14:25 19s
chunk done — 2026-08-10 14:25 0s
text_embed done — 2026-08-10 19:51 0s
keyframe done — 2026-08-10 14:25 1m 25s
ocr done — 2026-08-10 14:27 9s
frame_embed done — 2026-08-10 19:51 3s

Frames, and what the machine read

  • 0:03 #0 done2 line(s)

    shot 0·sharpness 658.4

    1. AlEngineer0.98
    2. EUROPE1.00

  • 0:08 #1 done2 line(s)

    shot 1·sharpness 829.1

    1. PRESENTINGSPONSOR1.00
    2. Google DeepMind1.00

  • 0:13 #2 done3 line(s)

    shot 2·sharpness 907.7

    1. PLATINUM SPONSORS0.98
    2. # Braintrust0.96
    3. WorkOS OpenAI0.95

  • 0:33 #3 done9 line(s)

    shot 3·sharpness 533.0

    1. AlEngineer0.99
    2. Agents1.00
    3. EUROPE1.00
    4. don't do1.00
    5. K0.65
    6. stand-ups1.00
    7. Building the post-engineering org0.99
    8. AlEngineer1.00
    9. EUROPE1.00

  • 0:49 #4 done16 line(s)

    shot 4·sharpness 1418.3

    1. ×C0.57
    2. docs.google.com/presentation/d/1qlqG-SNOtcfgclclWEWFV2G5mklx9wg8HBDtICxdsRl/edit?slide=id.g3d4d228785a_0_44#slide=id.g3d4d..0.99
    3. Agents1.00
    4. *★0.63
    5. AIE1.00
    6. don't do0.95
    7. ★1.00
    8. ★1.00
    9. ★1.00
    10. stand-ups1.00
    11. Building the post-engineering org0.99
    12. Mike Spitz0.99
    13. Google DeepMind0.99
    14. AlEngineer0.99
    15. EUROPE1.00
    16. 20261.00

  • 1:27 #5 done17 line(s)

    shot 5·sharpness 1409.5

    1. ×C0.60
    2. AX-GE·+0.62
    3. =docs.google.com/presentation/d/1qlqG-SNOtcfgclclWEWFV2G5mklx9wg8HBDtICxdsRl/edit?slide=id.g3d50c5dc8a1_0_...0.97
    4. Relaunch to update1.00
    5. 100m+1.00
    6. *★*0.53
    7. AIE1.00
    8. pageviews/yr1.00
    9. ★1.00
    10. ★1.00
    11. 9m+m0.94
    12. mock drafts started/yr1.00
    13. Braintrust1.00
    14. WorkOS OpenAI0.97
    15. AlEngineer0.99
    16. EUROPE1.00
    17. 20261.00

  • 1:48 #6 done20 line(s)

    shot 6·sharpness 1389.5

    1. MMM× ACSX@COS0.55
    2. GASX-E·+0.55
    3. docs.google.com/presentation/d/1qlqG-SNOtcfgclclWEWFV2G5mklx9wg8HBDtICxdsRI/edit?slide=id.g3d4d228785a_0..0.98
    4. ☆0.90
    5. Relaunch to update1.00
    6. From how do we help1.00
    7. engineers output more...1.00
    8. ***0.78
    9. AIE1.00
    10. ★1.00
    11. ★1.00
    12. ★1.00
    13. ★1.00
    14. ...to how do we make0.95
    15. agents faster?1.00
    16. AlEngineer0.96
    17. AlEngineer0.99
    18. EUROPE1.00
    19. EUROPE1.00
    20. 20261.00

  • 2:14 #7 skipped

    shot 7·duplicate of #5

  • 2:51 #8 done17 line(s)

    shot 8·sharpness 1363.5

    1. M×C0.58
    2. +0.93
    3. docs.google.com/presentation/d/1qlqG-SNOtcfgclclWEWFV2G5mklx9wg8HBDtICxdsRI/edit?slide=id.g3d4d228785a_0..0.98
    4. ☆0.87
    5. Relaunch to update1.00
    6. Results: deploy frequency1.00
    7. AIE1.00
    8. 25xm0.85
    9. ★1.00
    10. 5 per day vs 1 per week0.97
    11. ★1.00
    12. ★0.99
    13. Compared to team average0.99
    14. Engineering the future of Al0.99
    15. AlEngineer0.98
    16. EUROPE1.00
    17. 20261.00

  • 3:18 #9 done8 line(s)

    shot 9·sharpness 531.7

    1. Results: output per engineer1.00
    2. AlEngineer0.99
    3. EUROPE1.00
    4. 10x~0.88
    5. 2M1.00
    6. co0.74
    7. AlEngineer0.98
    8. EUROPE1.00

  • 3:32 #10 skipped

    shot 10·duplicate of #8

  • 3:59 #11 done63 line(s)

    shot 11·sharpness 1669.4

    1. MMM×S AGSS@COS0.56
    2. X-·+0.68
    3. docs.google.com/presentation/d/1qlqG-SNOtcfgclclWEWFV2G5mkIx9wg8HBDtiCxdsRl/edit?slide=id.g3d4d228785a_0..0.98
    4. ☆0.96
    5. Relaunch to update1.00
    6. Results: Al augmented engineering0.99
    7. 5 features shipped by 2 engineers over 7 weeks — with one engineer freed for a new product at week 3.60.99
    8. ★1.00
    9. ★1.00
    10. *★★0.61
    11. AIE1.00
    12. ★★0.74
    13. 16w1.00
    14. WITHOUT AI1.00
    15. 7w0.98
    16. WITH AI (ACTUAL)0.99
    17. 2.7×0.99
    18. CAPACITY MULTIPLIER0.99
    19. 8.6/100.96
    20. AVG QUALITY SCORE0.98
    21. ★1.00
    22. ★1.00
    23. ★1.00
    24. ★1.00
    25. 2 engineers, both fully occupled0.98
    26. 1 englneer freed at week 3.60.99
    27. 28 eng-weeks output from 10.2 Input0.99
    28. Across all 5 features1.00
    29. ACTUAL1.00
    30. With Al —7 weeks elapsed0.97
    31. Engineer A1.00
    32. MDS Multiplayer · 4w0.93
    33. My Drafts · 2w0.96
    34. Engineer B1.00
    35. BB Builder · 2w0.94
    36. BB MDS1.00
    37. → new product0.99
    38. 1w0.75
    39. 2w1.00
    40. 3w0.99
    41. 4w0.97
    42. 5w0.98
    43. 6w0.99
    44. 7w1.00
    45. COUNTERFACTUAL1.00
    46. Without Al — 16 weeks elapsed0.96
    47. Engineer A1.00
    48. LB·2W0.85
    49. BB Builder - 8w0.93
    50. MDS- 2w0.82
    51. My Drafts · 4w0.95
    52. Engineer B1.00
    53. MDS Multiplayer-12w0.99
    54. 2w0.97
    55. 8w0.95
    56. 10w0.99
    57. 12w0.97
    58. 14w0.93
    59. 16w0.96
    60. AlEngineer0.96
    61. AlEngineer0.98
    62. EUROPE1.00
    63. EUROPE1.00

  • 4:25 #12 skipped

    shot 12·duplicate of #11

  • 4:51 #13 done62 line(s)

    shot 13·sharpness 1672.2

    1. MMM× AGSXS@COS0.57
    2. SX+0.73
    3. docs.google.com/presentation/d/1qlqG-SNOtcfgclclWEWFV2G5mklIx9wg8HBDtICxdsRl/edit?slide=id.g3d4d228785a_0..0.99
    4. ☆0.97
    5. Relaunch to update1.00
    6. Results: Al augmented engineering0.99
    7. 5 features shipped by 2 engineers over 7 weeks — with one engineer freed for a new product at week 3.60.99
    8. *★★0.58
    9. AIE1.00
    10. ★1.00
    11. ★0.99
    12. 16w1.00
    13. WITHOUT AI1.00
    14. 7w1.00
    15. WITH AI (ACTUAL)1.00
    16. 2.7×0.97
    17. CAPACITY MULTIPLIER0.98
    18. 8.6/100.99
    19. AVG QUALITY SCORE0.97
    20. ★1.00
    21. ★1.00
    22. ★1.00
    23. ★1.00
    24. 2 englneers, both fully occupled0.96
    25. 1 engineer freed at week 3.60.97
    26. 28 eng-weeks output from 10.2 input0.98
    27. Across all 5 features1.00
    28. ACTUAL1.00
    29. With Al — 7 weeks elapsed0.94
    30. Engineer A1.00
    31. MDS Multiplayer ·4w0.95
    32. My Drafts · 2w0.96
    33. Engineer B1.00
    34. BB Builder · 2w0.92
    35. BB MDS1.00
    36. → new product0.96
    37. hw0.71
    38. 2w1.00
    39. 3w1.00
    40. 4w0.93
    41. 5w0.77
    42. 6w0.97
    43. COUNTERFACTUAL1.00
    44. Without Al — 16 weeks elapsed0.96
    45. Engineer A1.00
    46. LB·2W0.87
    47. BB Builder - 8w0.92
    48. MDS-2w0.94
    49. My Drafts · 4w0.93
    50. Engineer B1.00
    51. MDS Multiplayer·12w0.97
    52. 2w0.97
    53. 4w0.99
    54. 6w0.99
    55. 8w0.98
    56. 10w0.99
    57. 12w1.00
    58. 14w0.98
    59. 16w0.98
    60. Engineering the future of Al0.98
    61. AlEngineer1.00
    62. EUROPE1.00

  • 5:01 #14 skipped

    shot 14·duplicate of #8

  • 5:31 #15 done19 line(s)

    shot 15·sharpness 1344.0

    1. M×SAGSXCOS0.56
    2. ASXF-E·+0.62
    3. docs.google.com/presentation/d/1qlqG-SNOtcfgclclWEWFV2G5mklx9wg8HBDtICxdsRl/edit?slide=id.g3d50c5dc8a1_0_...0.98
    4. ☆0.95
    5. Relaunch to update1.00
    6. Our workflow1.00
    7. AIE1.00
    8. Ceremonies1.00
    9. Huddle Sessions1.00
    10. ★1.00
    11. ★1.00
    12. ★1.00
    13. ★1.00
    14. Spec1.00
    15. Tickets + PRs1.00
    16. Engineering the future of Al0.99
    17. AlEngineer0.99
    18. EUROPE1.00
    19. 20261.00

  • 6:26 #16 skipped

    shot 16·duplicate of #6

  • 6:50 #17 skipped

    shot 17·duplicate of #9

  • 7:03 #18 skipped

    shot 18·duplicate of #8

  • 7:34 #19 skipped

    shot 19·duplicate of #5

  • 8:15 #20 skipped

    shot 20·duplicate of #5

  • 8:25 #21 skipped

    shot 21·duplicate of #5

  • 9:03 #22 done28 line(s)

    shot 22·sharpness 1508.9

    1. MMM×SAGSXS@OMS0.54
    2. SASXTE+0.70
    3. docs.google.com/presentation/d/1qlqG-SNOtcfgclclWEWFV2G5mklx9wg8HBDtICxdsRl/edit?slide=id.g3d50c5dc8a1_0_..0.99
    4. ☆0.98
    5. Relaunch to update0.98
    6. Where to start?: with your best1.00
    7. ★1.00
    8. AIE1.00
    9. ★0.99
    10. ★1.00
    11. 11.00
    12. 21.00
    13. 31.00
    14. ★1.00
    15. ★1.00
    16. ★1.00
    17. Pick engineers1.00
    18. Scale slowly0.99
    19. Experiment in0.99
    20. with deep, broad0.99
    21. non-critical1.00
    22. system knowledge1.00
    23. systems1.00
    24. AlEngineer0.95
    25. AlEngineer0.99
    26. EUROPE1.00
    27. EUROPE0.98
    28. 20261.00

  • 9:31 #23 done7 line(s)

    shot 23·sharpness 542.4

    1. AlEngineer0.98
    2. "Not ever0.95
    3. EUROPE1.00
    4. can drive1.00
    5. sports ca1.00
    6. AlEngineer0.97
    7. EUROPE1.00

Transcript

187 cues· 2,897 words· 15,570 chars

  1. 0:15 All right, so I am in the slot before the snacks.
  2. 0:19 But I'm here to speak about the post-engineering org and a case study we've been doing at PFF.
  3. 0:26 Started in January and we finished it in March.
  4. 0:29 PFF is a sports data company.
  5. 0:31 We help NFL and NCAA teams figure out what they should be doing.
  6. 0:37 And we also have a consumer arm which does fantasy football
  7. 0:43 Sport betting, and for those of you who know American football, there's a draft happening end of the month, and so it allows people to basically play as one of these teams.
  8. 0:52 We're a fully distributed engineering team, so we've got engineers in India, we've got engineers in Spain, and all the states in America.
  9. 1:01 But first let's go into some kind of stats.
  10. 1:04 So we have 100 million page views annually.
  11. 1:07 We have nine million drafts happen on an annual basis.
  12. 1:10 It's a fairly popular tool.
  13. 1:14 And the issue that we had was we were 200 employees, around 20 engineers.
  14. 1:20 And we were falling behind competitors.
  15. 1:24 We were focusing on some sport betting stuff, but a lot of our consumers were actually interested in this.
  16. 1:29 And then with Cloud Opus coming out, I started experimenting around November on a personal level.
  17. 1:36 And then it got spun out to two engineers.
  18. 1:40 One of them was our strongest front-end engineer, and the other one was one of our strongest full-stack engineer.
  19. 1:48 So really, the question I was asking was, instead of figuring out how we can help engineers go and output more, how do we help make the agents quicker?
  20. 2:00 So I think if we all think, historically, software engineering, you've got the agile manifesto, software craftsmanship.
  21. 2:08 You've got a lot of good perks and benefits for engineers, from foosball tables to sleeping pods to pretty much every perk that most other industries really don't have.
  22. 2:17 And it's because we're the bottleneck.
  23. 2:19 And if companies are able to go and optimize for that, it gives them a lot of benefit.
  24. 2:24 But now things are changing.
  25. 2:27 So I just want to go straight into how the case study ended up.
  26. 2:30 So we had 25 times more deploys.
  27. 2:32 So the two engineers were deploying five times
  28. 2:40 five times every, we're deploying five times every day.
  29. 2:44 And the other team was a team of around 10 engineers.
  30. 2:47 They're doing it pretty much one deploy every five days.
  31. 2:51 There is a big obvious caveat here, right?
  32. 2:52 Small engineering teams are always gonna be quicker than the big ones.
  33. 2:56 So there is an element of that multiplier that is just from it being a small engineering team.
  34. 3:02 However, that smaller Tiger team still had to coordinate all of those daily, still had to go and coordinate all of those deploys with the bigger team, so there was still that kind of issue that was happening.
  35. 3:16 And this one, this is always hard, how do you validate like the output is actually helpful, the number of PRs isn't helpful, the amount of code isn't helpful.
  36. 3:24 So we basically blended the number of tickets with the code complexity, and we found that their output was 10x.
  37. 3:35 Got another slide which might look a little bit hectic, but just bear with me.
  38. 3:40 So these are the features that we went and constructed with those two engineers that took them under a two month.
  39. 3:47 If we're gonna do this before, we're estimating it's gonna take four months.
  40. 3:52 The big thing you can probably have a look in the top half is one of the engineers gets unblocked in under a month and can then start building other stuff.
  41. 4:01 Whereas in the old way, they're both blocked for three months.
  42. 4:05 And so you get this thing where you now have a compounding increase where it's not just faster for that stuff, but you're now able to do a lot more stuff than you were able to do before.
  43. 4:17 And the one thing that I really want to go and basically go and highlight is it doesn't matter if the output's more.
  44. 4:24 It doesn't matter if the number of deployments are higher.
  45. 4:27 What really matters is basically if the customers are happy.
  46. 4:32 And so we did statistically significant surveys, and the average quality score was 8.6 out of 10.
  47. 4:41 What was interesting was before AI, we would probably average seven, seven and a half.
  48. 4:47 So we weren't really delivering what the customers had been interested in.
  49. 4:53 So...
  50. 4:55 Scrum did not survive we were we were basically having a look not just on an engineering front and all of the delivery gains from that but also from a process standpoint and so

Chapters

  1. 0:00 Introduction to the case study at PFF
  2. 1:47 The shift: optimizing for agent speed rather than engineer speed
  3. 2:28 Results: 25x increase in deployment frequency
  4. 3:16 Measuring success through ticket complexity and customer satisfaction
  5. 4:53 Dismantling Scrum and traditional development processes
  6. 5:58 The new development workflow (Spec → LDD → Ticket → PR)
  7. 6:51 Eliminating coordination overhead (no sprint planning or standups)
  8. 8:28 Best practices for implementation and team selection
  9. 10:09 Utilizing agents for deterministic tasks and code reviews
  10. 12:00 Viewing the engineering lifecycle as a factory
  11. 13:06 Automated QA and the future of self-healing systems
  12. 14:10 Where human oversight remains essential
  13. 14:48 Strategic advice for scaling AI-driven engineering

Open at this second