vidtheque.

read-only demo

Videos iUWwcG-C8OU

Why Can't Anyone Answer Questions About the Business? — Garrett Galow, WorkOS

index_state ready data_status ok

AI Engineer· published 2026-06-11· 0:19:05· en-US· indexed 2026-08-11 10:45

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 0:41, 1 of 1 keyframes kept
  5. Shot 4, 0:41 to 1:06, 1 of 1 keyframes kept
  6. Shot 5, 1:06 to 1:31, 1 of 1 keyframes kept
  7. Shot 6, 1:31 to 1:56, 1 of 1 keyframes kept
  8. Shot 7, 1:56 to 2:22, 0 of 1 keyframes kept
  9. Shot 8, 2:22 to 2:47, 0 of 1 keyframes kept
  10. Shot 9, 2:47 to 2:50, 1 of 1 keyframes kept
  11. Shot 10, 2:50 to 2:53, 1 of 1 keyframes kept
  12. Shot 11, 2:53 to 3:19, 0 of 1 keyframes kept
  13. Shot 12, 3:19 to 3:46, 0 of 1 keyframes kept
  14. Shot 13, 3:46 to 4:12, 1 of 1 keyframes kept
  15. Shot 14, 4:12 to 4:47, 1 of 1 keyframes kept
  16. Shot 15, 4:47 to 5:21, 0 of 1 keyframes kept
  17. Shot 16, 5:21 to 5:48, 0 of 1 keyframes kept
  18. Shot 17, 5:48 to 6:16, 0 of 1 keyframes kept
  19. Shot 18, 6:16 to 6:43, 0 of 1 keyframes kept
  20. Shot 19, 6:43 to 7:10, 0 of 1 keyframes kept
  21. Shot 20, 7:10 to 7:37, 0 of 1 keyframes kept
  22. Shot 21, 7:37 to 8:04, 1 of 1 keyframes kept
  23. Shot 22, 8:04 to 8:31, 1 of 1 keyframes kept
  24. Shot 23, 8:31 to 8:58, 0 of 1 keyframes kept
  25. Shot 24, 8:58 to 9:25, 0 of 1 keyframes kept
  26. Shot 25, 9:25 to 9:44, 1 of 1 keyframes kept
  27. Shot 26, 9:44 to 9:49, 1 of 1 keyframes kept
  28. Shot 27, 9:49 to 9:53, 0 of 1 keyframes kept
  29. Shot 28, 9:53 to 9:54, 1 of 1 keyframes kept
  30. Shot 29, 9:54 to 10:21, 0 of 1 keyframes kept
  31. Shot 30, 10:21 to 10:48, 0 of 1 keyframes kept
  32. Shot 31, 10:48 to 11:14, 0 of 1 keyframes kept
  33. Shot 32, 11:14 to 11:40, 0 of 1 keyframes kept
  34. Shot 33, 11:40 to 12:06, 0 of 1 keyframes kept
  35. Shot 34, 12:06 to 12:32, 0 of 1 keyframes kept
  36. Shot 35, 12:32 to 12:58, 0 of 1 keyframes kept
  37. Shot 36, 12:58 to 13:24, 0 of 1 keyframes kept
  38. Shot 37, 13:24 to 13:50, 0 of 1 keyframes kept
  39. Shot 38, 13:50 to 14:16, 0 of 1 keyframes kept
  40. Shot 39, 14:16 to 14:42, 0 of 1 keyframes kept
  41. Shot 40, 14:42 to 15:08, 0 of 1 keyframes kept
  42. Shot 41, 15:08 to 15:34, 1 of 1 keyframes kept
  43. Shot 42, 15:34 to 16:00, 0 of 1 keyframes kept
  44. Shot 43, 16:00 to 16:26, 0 of 1 keyframes kept
  45. Shot 44, 16:26 to 16:52, 0 of 1 keyframes kept
  46. Shot 45, 16:52 to 17:19, 0 of 1 keyframes kept
  47. Shot 46, 17:19 to 17:45, 0 of 1 keyframes kept
  48. Shot 47, 17:45 to 18:11, 0 of 1 keyframes kept
  49. Shot 48, 18:11 to 18:50, 1 of 1 keyframes kept
  50. Shot 49, 18:50 to 19:05, 1 of 1 keyframes kept

50 shot(s).

keyframes kept every frame deduplicated

What was stored

cues
259
whisperx 259
chunks
34
from 259 cues
keyframes
19
kept of 50 captured
frames with text
19
748 lines read
chapters
0
from the source metadata
keyframe bytes
5.2 MB
word timings on 259 cues

Provenance

Each pipeline stage, its state and the model that produced it
stage state model started took
fetch done — 2026-08-11 10:41 1m 33s
stt done — 2026-08-11 10:43 20s
chunk done — 2026-08-11 10:43 0s
text_embed done — 2026-08-11 10:43 1s
keyframe done — 2026-08-11 10:43 1m 35s
ocr done — 2026-08-11 10:45 11s
frame_embed done — 2026-08-11 10:45 4s

Frames, and what the machine read

  • 0:03 #0 done2 line(s)

    shot 0·sharpness 662.1

    1. AlEngineer0.98
    2. EUROPE1.00

  • 0:08 #1 done2 line(s)

    shot 1·sharpness 814.9

    1. PRESENTINGSPONSOR1.00
    2. Google DeepMind1.00

  • 0:12 #2 done3 line(s)

    shot 2·sharpness 905.6

    1. PLATINUM SPONSORS0.98
    2. # Braintrust0.96
    3. WorkOS OpenAI0.94

  • 0:22 #3 done3 line(s)

    shot 3·sharpness 475.4

    1. Hi Garrett,0.98
    2. What can we help with today?0.99
    3. AIEr0.86

  • 0:56 #4 done19 line(s)

    shot 4·sharpness 1841.3

    1. studio.workos.dev0.99
    2. ② 日0.56
    3. WorkOS Studio1.00
    4. New Space1.00
    5. 88 Browse0.97
    6. Settings0.96
    7. Spaces1.00
    8. Radar lookup1.00
    9. Content Attribution0.99
    10. Changelog updates0.99
    11. Connect Pricing Simulator1.00
    12. AIE1.00
    13. Hi Garrett,0.98
    14. What can we help with today?1.00
    15. ★1.00
    16. ★1.00
    17. Ask a question or describe what you want to build0.99
    18. Garrett Galow0.96
    19. Engineering the future of Al0.99

  • 1:21 #5 done22 line(s)

    shot 5·sharpness 1923.6

    1. studio.workos.dev0.97
    2. ☐ 日0.58
    3. WorkOS Studio0.97
    4. New Space1.00
    5. 88 Browse0.97
    6. Settings0.95
    7. Spaces1.00
    8. Radar lookup1.00
    9. Content Attribution1.00
    10. Changelog updates0.99
    11. Connect Pricing Simulator1.00
    12. AIE1.00
    13. Hi Garrett,1.00
    14. ★1.00
    15. What can we help with today?1.00
    16. ★1.00
    17. ★1.00
    18. ★1.00
    19. Ask a question or describe what you want to build1.00
    20. Garrett Galow1.00
    21. Braintrust1.00
    22. WorkOS OpenAI0.94

  • 1:53 #6 done27 line(s)

    shot 6·sharpness 2019.8

    1. node-id=0-18p=f&t=dn7ufverrev140oe-00.98
    2. 甲Studio0.91
    3. 6.03000.91
    4. 1. The Current Way0.97
    5. ★1.00
    6. ★1.00
    7. AIE1.00
    8. business question1.00
    9. Someone has a0.96
    10. Find the person1.00
    11. who knows SQL0.99
    12. Explain the question1.00
    13. + all the context0.97
    14. Wait1.00
    15. Get an answer1.00
    16. Verify?0.99
    17. Lose it forever0.98
    18. Share in Slack1.00
    19. ★1.00
    20. ★1.00
    21. ★1.00
    22. ★1.00
    23. ★1.00
    24. Nope. Try again.0.99
    25. 800.75
    26. AlEngineer0.97
    27. EUROPE1.00

  • 1:59 #7 skipped

    shot 7·duplicate of #6

  • 2:25 #8 skipped

    shot 8·duplicate of #6

  • 2:49 #9 done5 line(s)

    shot 9·sharpness 2364.4

    1. AIE1.00
    2. ★1.00
    3. ★1.00
    4. zoom0.86
    5. Engineering the future of Al0.99

  • 2:51 #10 done6 line(s)

    shot 10·sharpness 1787.8

    1. **★0.67
    2. AIE1.00
    3. ★1.00
    4. ★1.00
    5. 40.99
    6. Engineering the future of Al1.00

  • 3:11 #11 skipped

    shot 11·duplicate of #4

  • 3:22 #12 skipped

    shot 12·duplicate of #4

  • 4:06 #13 done70 line(s)

    shot 13·sharpness 2613.6

    1. ☐ 日0.52
    2. WorkOS0.92
    3. Stellar Mentor ReportsStellar Mentor Report- 2026-0...0.97
    4. Edited Apr 1 è Share0.89
    5. ☆…0.97
    6. Home1.00
    7. Q0.99
    8. Upcoming events1.00
    9. AIE Expo | Booth...1:30 - 4:30 PM0.90
    10. Expo Session ...2:05 - 2:23 PM0.89
    11. Austin Free-Net Tent...3- 4 PM0.94
    12. anthropic sessio... 3:10 - 3:40 PM0.97
    13. Add coverAdd verificationAdd comment0.99
    14. All Hands0.94
    15. 5-69PM0.87
    16. Stellar Mentor Report —2026-04-0.98
    17. Open Meetings0.98
    18. 011.00
    19. Recents0.99
    20. AIE1.00
    21. ★1.00
    22. ★1.00
    23. Stellar Mentor Reports1.00
    24. All Hands - April 3, 20260.93
    25. AlI Hands0.93
    26. Lookback window: Start of business Mon Mar 31, 2026 → Wed Apr 1, 2026 9:30 AM CT0.99
    27. ★1.00
    28. All Hands - April 10, 20260.91
    29. Slack Channel Summaries1.00
    30. ★1.00
    31. ★1.00
    32. ★1.00
    33. Product Manager Interview G...0.98
    34. Favorites1.00
    35. #team-authkit0.97
    36. Product Project Lifecycle0.97
    37. Per-org branding is the dominant theme. Maxwell shared customer feedback that per-tenant1.00
    38. UI branding is the single biggest reason customers avoid hosted AuthKit. MG, Garrett, and0.98
    39. Agents Beta0.91
    40. Randall discussed an "AuthKit for Platforms" enterprise SKU. Customer interviews confirm0.99
    41. Stellar Mentor0.97
    42. demand for per-org logos, colors, email customization, and org-slug subdomain routing.[1]1.00
    43. Product Owner Sync1.00
    44. Key threads:1.00
    45. More1.00
    46. • ToS/EULA checkbox demand — MG flagged a customer request for a EULA acceptance0.99
    47. Private0.99
    48. Stellar Mentor Reports0.99
    49. xaa-talk-outline0.97
    50. • Waitlist feature — MG noted waitlist has been on the backlog (Clerk has it) and is relevant0.98
    51. for Mira Murati's startup. Asked if the AuthKit team can quick-build it. [3]0.99
    52. FEAT-835 in Linear with the demand signal.[2]0.98
    53. checkbox at signup. Glean surfaced 9 distinct asks across 8 customer orgs. Alex updated0.99
    54. EPD staff @March 23, 2026 1..0.95
    55. • Multi-app invite routing gap — Heather flagged that multi-app support doesn't yet support0.99
    56. New page0.95
    57. per-application invitation URLs. Michael Hadley confirmed invites only work with the "default0.99
    58. AIE London - XAA Breakout s...0.96
    59. app" — per-app invite support still needs to be built.[4]0.96
    60. More1.00
    61. AlE Expo | Booth Shift B0.94
    62. 1:30 - 4:30 PM 3h0.88
    63. • Per-org session/token length controls — Customer (Nikita Lilichenko) asking for per-0.98
    64. organization session and access token length controls as an urgent priority. [5]1.00
    65. New chat x00.88
    66. Transcribe1.00
    67. • Dogfooding - Kesin investigated woRKOS_CLIENT_ID env variable for the API service. Self-0.97
    68. resolved.[6]1.00
    69. Al Engineer0.92
    70. EUROPE1.00

  • 4:43 #14 done32 line(s)

    shot 14·sharpness 2323.2

    1. Studio0.93
    2. 603000.88
    3. User types a question0.99
    4. Next.js API0.99
    5. LangGraph Agent...0.97
    6. Update Widget1.00
    7. Answer or1.00
    8. POST /api/chat...0.97
    9. or asks via Slack0.97
    10. Convex1.00
    11. ★1.00
    12. AIE1.00
    13. ★1.00
    14. integrationContexts1.00
    15. Guidance Layer1.00
    16. Agent Tools1.00
    17. conversations1.00
    18. messages1.00
    19. widgets...0.91
    20. ★1.00
    21. ★1.00
    22. ★1.00
    23. org-specific rules1.00
    24. schema + naming0.99
    25. Integration Proxy1.00
    26. /api/integrations/...0.98
    27. Data Sources1.00
    28. Snowflake1.00
    29. Linear (REST)0.98
    30. Notion (SDK)0.99
    31. 800.78
    32. Engineering the future of Al0.98

  • 4:54 #15 skipped

    shot 15·duplicate of #14

  • 5:30 #16 skipped

    shot 16·duplicate of #4

  • 5:59 #17 skipped

    shot 17·duplicate of #6

  • 6:24 #18 skipped

    shot 18·duplicate of #6

  • 6:56 #19 skipped

    shot 19·duplicate of #4

  • 7:18 #20 skipped

    shot 20·duplicate of #4

  • 7:48 #21 done55 line(s)

    shot 21·sharpness 2148.1

    1. studio.workos.dev/t/thread_2e1fs212-b0b6-4d36-9fec-a00tct98o5fc0.95
    2. WorkOS Studio0.98
    3. Radar User Lookup1.00
    4. ①Share0.93
    5. • Fixed email lookup - now uses emai l_hash =0.95
    6. New Space1.00
    7. MD5(L0WER(email)) instead of plaintext matching0.98
    8. • Detections query — joins through0.97
    9. 88 Browse0.94
    10. Radar User Lookup1.00
    11. radar_auth_attempt_id to find detections linked1.00
    12. v31.00
    13. to matching attempts1.00
    14. Settings1.00
    15. • Collapsible customer sections — click a customer0.98
    16. header to expand/collapse1.00
    17. Spaces1.00
    18. Radar User Lookup0.99
    19. • Expandable detection rows — click any attempt with0.98
    20. can you tell me which piece...0.99
    21. Search by email to view auth attempts and detections across customers0.99
    22. detections to see inline details (action, control,0.99
    23. reason, challenge info)1.00
    24. Radar lookup0.97
    25. Enter user email..1.00
    26. Search1.00
    27. • Added "Blocked" stat and grant type column for0.98
    28. more context1.00
    29. ★1.00
    30. ★1.00
    31. AIE1.00
    32. ★1.00
    33. Content Attribution1.00
    34. Changelog updates1.00
    35. • Color-coded rows — blocked attempts get a red0.96
    36. background tint1.00
    37. ★1.00
    38. ★1.00
    39. ★1.00
    40. ★1.00
    41. ★1.00
    42. Connect Pricing Simulator1.00
    43. can you tweak the column width on type so the text1.00
    44. is on one row1.00
    45. ✓ Thinking...0.95
    46. Radar User Lookup Completed1.00
    47. Fixed the Type column — added whiteSpace:0.98
    48. 'nowrap' and minWidth: 70 to the <td> , plus0.98
    49. whiteSpace: 'nowrap' and display: 'inline-1.00
    50. block* on the Sign Up / Sign In badge spans so they'll0.98
    51. always render on a single line.0.98
    52. Ask a question or describe what you want to1.00
    53. build1.00
    54. Garrett Galow1.00
    55. Google DeepMind1.00

  • 8:28 #22 done123 line(s)

    shot 22·sharpness 2382.9

    1. 日0.53
    2. studio.workos.dev/t/thread_2e1f4212-b0b6-4d36-9fec-a00tct98e510.94
    3. 日0.84
    4. WorkOS Studio0.97
    5. Radar User Lookup0.98
    6. ①Share0.95
    7. Here's what I built:0.97
    8. New Space1.00
    9. Radar User Lookup Dashboard0.99
    10. 88Browse0.95
    11. Radar User Lookup1.00
    12. How it works:1.00
    13. Settings1.00
    14. v3~0.80
    15. 1. Enter an email in the search bar and hit Enter or click0.97
    16. Search1.00
    17. Spaces1.00
    18. Radar User Lookup1.00
    19. 2. The widget queries both RADAR_AUTH_ATTEMPTS1.00
    20. and RADAR_DETECTIONS tables, joining through to0.99
    21. can you tell me which piece....0.98
    22. Search by email to view auth attempts and detections across customers0.99
    23. the teams table to identify which customer each1.00
    24. record belongs to1.00
    25. Radar lookup0.97
    26. [email protected]0.99
    27. Search1.00
    28. Key features:1.00
    29. ★1.00
    30. ★1.00
    31. Content Attribution1.00
    32. ★1.00
    33. ★1.00
    34. ★1.00
    35. AIE1.00
    36. ★1.00
    37. ★1.00
    38. ★1.00
    39. ★1.00
    40. Changelog updates1.00
    41. Connect Pricing Simulator1.00
    42. Customers1.00
    43. 61.00
    44. Auth Attempts1.00
    45. 261.00
    46. Detections1.00
    47. • Summary stats at the top showing total customers,0.98
    48. • Grouped by customer — each customer (team) gets0.98
    49. auth attempts, and detections at a glance1.00
    50. in the headey0.96
    51. its own card with attempt count and detection count0.99
    52. • Expandable rows — click any attempt row that has0.99
    53. Cursortean_01656W3C7CY90901MRP4E99N130.97
    54. 10 attempts1.00
    55. 1 detection1.00
    56. detection details inline (action, control, reason,1.00
    57. detections (indicated by) to expand and see the1.00
    58. TIME1.00
    59. STATUS1.00
    60. TYPE1.00
    61. GRANT1.00
    62. IP ADDRESS1.00
    63. LOC0.81
    64. challenge type, device)0.99
    65. Feb 20, 2026, 03:50:54 PM1.00
    66. successtul0.91
    67. Sign In0.98
    68. password1.00
    69. 2a09:bac2:bab3:2628::3cd:190.97
    70. Dal1.00
    71. • Color-coded badges for attempt status0.99
    72. (succeeded/failed), type (sign up/sign in), action1.00
    73. Feb 17, 2026, 05:38:19 AM0.99
    74. successful0.99
    75. Sign In0.99
    76. password1.00
    77. 2a09:bac3:bdd0:15d7::22d:1240.97
    78. Aus0.78
    79. (allow/deny/challenge/flag), and bot scores (green <1.00
    80. 40, yellow 40-69, red ≥ 70)0.98
    81. Jul 18, 2025, 04:34:49 PM1.00
    82. successtul0.97
    83. Sign In0.99
    84. password1.00
    85. 2600:1708:280:7620:ad8c:c69d:9707:47ed0.98
    86. Aut0.95
    87. • Challenge tracking — shows whether a challenge was0.97
    88. issued and if it was corrd0.98
    89. Jun 12, 2025, 09:09:55 PM1.00
    90. successful0.97
    91. Sign In1.00
    92. password1.00
    93. 104,13.169.60.96
    94. Aus0.88
    95. Mar 19, 2025, 04:29:06 PM0.99
    96. attempted1.00
    97. Sign Up1.00
    98. password1.00
    99. 104.13.169.60.96
    100. Aut0.85
    101. Ask a question or describe what you want to0.99
    102. Mar 12, 2025, 02:12:09 AM1.00
    103. successful0.99
    104. Sign In1.00
    105. password1.00
    106. 2600:1708:280:7620:6@c9:6688:5a86:c88a0.97
    107. Aut0.86
    108. build1.00
    109. Mar 7, 2025, 09:44:55 PM1.00
    110. successful0.99
    111. Sign In0.99
    112. password1.00
    113. 104,13.169.60.98
    114. Aus0.94
    115. Garrett Galow1.00
    116. Mar 5, 2025, 04:19:23 PM1.00
    117. attempted0.95
    118. Sign Up1.00
    119. password1.00
    120. 104,13.169.60.95
    121. Aut0.99
    122. AlEngineer0.96
    123. EUROPE1.00

  • 8:50 #23 skipped

    shot 23·duplicate of #6

Transcript

259 cues· 3,372 words· 18,000 chars

  1. 0:14 Good afternoon, everyone.
  2. 0:15 I know it's the last day.
  3. 0:17 Hopefully, you're still holding on, not too tired of talking about AI yet.
  4. 0:20 Yeah, today's going to be a little different.
  5. 0:24 So I work for WorkOS.
  6. 0:25 My name's Garrett.
  7. 0:26 I run product for the team.
  8. 0:28 I'm not exactly talking about our product today.
  9. 0:31 So I'll do like 10 seconds about WorkOS just to get that out of the way since my company cares about that.
  10. 0:37 We do enterprise platform features or developer platform.
  11. 0:41 The quick and easy is if you've ever logged into Cursor,
  12. 0:44 if you used WorkOS, whether that was username, password, or you went through your enterprise IDP.
  13. 0:49 We power enterprise platform features for the likes of Cursor, Anthropic, OpenAI.
  14. 0:55 Today, though, I'm going to talk about something a little bit different.
  15. 0:56 I'm going to talk about how we operate internally and things that we've built to make ourselves more productive.
  16. 1:02 So I imagine most of you are probably engineers or on the technical side.
  17. 1:09 You might have questions about things about your company, about how customers are using the product, how things are working.
  18. 1:15 Your go-to-market teams, your support teams definitely have questions about how customers are using your product, trying to figure out answers to questions.
  19. 1:23 You might have retool inside your company.
  20. 1:25 You might have built dashboards and things like that.
  21. 1:29 Of course, those can be fairly rigid.
  22. 1:32 You build a very specific thing.
  23. 1:34 Someone comes and says, oh, actually, but I need this extra bit of data.
  24. 1:36 I need to answer this different question that the dashboard doesn't answer.
  25. 1:39 And so either you go and build that, you change it.
  26. 1:43 We kind of see a workflow that looks kind of like this, where someone has a question, often about the business.
  27. 1:49 They may not be technical enough to go answer it themselves.
  28. 1:53 They often need something like SQL or someone that has access to the data.
  29. 1:56 They have to explain their question, why they need it answered, the context to answer it.
  30. 2:01 They wait.
  31. 2:02 Someone like you has to go answer the question, provide that data back to them.
  32. 2:06 Did you actually answer the question?
  33. 2:07 Did you provide enough detail?
  34. 2:08 Oh, no, that's great, but I actually need the next layer deeper.
  35. 2:11 Got to go back and forth.
  36. 2:13 You probably share that in Slack.
  37. 2:15 It's sort of a one-off.
  38. 2:16 Doesn't really scale very well.
  39. 2:19 We had this problem.
  40. 2:20 If you didn't, we have this problem every day.
  41. 2:23 And so we built a tool called Studio that serves as kind of an internal workspace where people can answer questions and build these kind of apps or dashboards themselves.
  42. 2:33 So I'm gonna show you both kind of what it looks like to build this out, also show you a few examples of some of the tools that we use every day inside of Studio.
  43. 2:40 And then I'll talk a little bit about how it works under the covers.
  44. 2:44 So I don't...
  45. 2:48 get this completely wrong.
  46. 2:50 I have a little prompt here already.
  47. 2:54 So a common thing we have, we do lots of marketing.
  48. 2:59 We're doing podcast advertisements.
  49. 3:00 We're doing Google Ads.
  50. 3:01 We're getting people to come to WorkOS site, whether that's our blog, our docs, or our marketing site.

Open at this second