vidtheque.

read-only demo

Videos vAIDdLKB6-w

A Piece of Pi: Embedding The OpenClaw Coding Agent In Your Product — Matthias Luebken, Tavon

index_state ready data_status ok

AI Engineer· published 2026-05-11· 0:20:42· en-US· indexed 2026-08-10 19:56

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 0:16, 1 of 1 keyframes kept
  5. Shot 4, 0:16 to 0:17, 1 of 1 keyframes kept
  6. Shot 5, 0:17 to 0:18, 1 of 1 keyframes kept
  7. Shot 6, 0:18 to 0:36, 1 of 1 keyframes kept
  8. Shot 7, 0:36 to 0:43, 1 of 1 keyframes kept
  9. Shot 8, 0:43 to 0:53, 1 of 1 keyframes kept
  10. Shot 9, 0:53 to 1:08, 1 of 1 keyframes kept
  11. Shot 10, 1:08 to 1:12, 1 of 1 keyframes kept
  12. Shot 11, 1:12 to 1:23, 1 of 1 keyframes kept
  13. Shot 12, 1:23 to 1:30, 1 of 1 keyframes kept
  14. Shot 13, 1:30 to 1:54, 1 of 1 keyframes kept
  15. Shot 14, 1:54 to 2:19, 1 of 1 keyframes kept
  16. Shot 15, 2:19 to 2:45, 1 of 1 keyframes kept
  17. Shot 16, 2:45 to 3:11, 1 of 1 keyframes kept
  18. Shot 17, 3:11 to 3:37, 1 of 1 keyframes kept
  19. Shot 18, 3:37 to 4:04, 1 of 1 keyframes kept
  20. Shot 19, 4:04 to 4:18, 1 of 1 keyframes kept
  21. Shot 20, 4:18 to 4:23, 1 of 1 keyframes kept
  22. Shot 21, 4:23 to 4:59, 1 of 1 keyframes kept
  23. Shot 22, 4:59 to 5:10, 1 of 1 keyframes kept
  24. Shot 23, 5:10 to 5:34, 1 of 1 keyframes kept
  25. Shot 24, 5:34 to 6:14, 1 of 1 keyframes kept
  26. Shot 25, 6:14 to 6:22, 1 of 1 keyframes kept
  27. Shot 26, 6:22 to 6:28, 1 of 1 keyframes kept
  28. Shot 27, 6:28 to 6:36, 1 of 1 keyframes kept
  29. Shot 28, 6:36 to 7:01, 1 of 1 keyframes kept
  30. Shot 29, 7:01 to 7:25, 1 of 1 keyframes kept
  31. Shot 30, 7:25 to 7:38, 1 of 1 keyframes kept
  32. Shot 31, 7:38 to 8:13, 1 of 1 keyframes kept
  33. Shot 32, 8:13 to 8:30, 1 of 1 keyframes kept
  34. Shot 33, 8:30 to 8:43, 1 of 1 keyframes kept
  35. Shot 34, 8:43 to 9:10, 1 of 1 keyframes kept
  36. Shot 35, 9:10 to 9:37, 0 of 1 keyframes kept
  37. Shot 36, 9:37 to 10:04, 1 of 1 keyframes kept
  38. Shot 37, 10:04 to 10:08, 1 of 1 keyframes kept
  39. Shot 38, 10:08 to 10:11, 1 of 1 keyframes kept
  40. Shot 39, 10:11 to 10:27, 1 of 1 keyframes kept
  41. Shot 40, 10:27 to 10:55, 1 of 1 keyframes kept
  42. Shot 41, 10:55 to 11:23, 0 of 1 keyframes kept
  43. Shot 42, 11:23 to 11:39, 1 of 1 keyframes kept
  44. Shot 43, 11:39 to 11:46, 1 of 1 keyframes kept
  45. Shot 44, 11:46 to 12:19, 1 of 1 keyframes kept
  46. Shot 45, 12:19 to 12:26, 1 of 1 keyframes kept
  47. Shot 46, 12:26 to 12:32, 1 of 1 keyframes kept
  48. Shot 47, 12:32 to 12:46, 0 of 1 keyframes kept
  49. Shot 48, 12:46 to 13:20, 1 of 1 keyframes kept
  50. Shot 49, 13:20 to 14:01, 1 of 1 keyframes kept
  51. Shot 50, 14:01 to 14:10, 1 of 1 keyframes kept
  52. Shot 51, 14:10 to 14:15, 1 of 1 keyframes kept
  53. Shot 52, 14:15 to 14:32, 1 of 1 keyframes kept
  54. Shot 53, 14:32 to 14:46, 1 of 1 keyframes kept
  55. Shot 54, 14:46 to 15:19, 1 of 1 keyframes kept
  56. Shot 55, 15:19 to 15:52, 1 of 1 keyframes kept
  57. Shot 56, 15:52 to 16:05, 1 of 1 keyframes kept
  58. Shot 57, 16:05 to 16:37, 1 of 1 keyframes kept
  59. Shot 58, 16:37 to 17:09, 1 of 1 keyframes kept
  60. Shot 59, 17:09 to 17:42, 0 of 1 keyframes kept
  61. Shot 60, 17:42 to 18:11, 1 of 1 keyframes kept
  62. Shot 61, 18:11 to 18:22, 1 of 1 keyframes kept
  63. Shot 62, 18:22 to 18:48, 1 of 1 keyframes kept
  64. Shot 63, 18:48 to 19:17, 1 of 1 keyframes kept
  65. Shot 64, 19:17 to 19:55, 1 of 1 keyframes kept
  66. Shot 65, 19:55 to 20:24, 1 of 1 keyframes kept
  67. Shot 66, 20:24 to 20:27, 1 of 1 keyframes kept
  68. Shot 67, 20:27 to 20:40, 1 of 1 keyframes kept
  69. Shot 68, 20:40 to 20:41, 0 of 1 keyframes kept

69 shot(s).

keyframes kept every frame deduplicated

What was stored

cues
235
whisperx 235
chunks
36
from 235 cues
keyframes
64
kept of 69 captured
frames with text
64
1,675 lines read
chapters
12
from the source metadata
keyframe bytes
7.0 MB
word timings on 235 cues

Provenance

Each pipeline stage, its state and the model that produced it
stage state model started took
fetch done — 2026-08-10 19:00 2m 13s
stt done — 2026-08-10 19:02 22s
chunk done — 2026-08-10 19:03 0s
text_embed done — 2026-08-10 19:56 1s
keyframe done — 2026-08-10 19:03 1m 59s
ocr done — 2026-08-10 19:05 40s
frame_embed done — 2026-08-10 19:56 11s

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 907.6

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

  • 0:15 #3 done30 line(s)

    shot 3·sharpness 1038.2

    1. AIEngineer0.95
    2. OpenAI0.93
    3. AlEngineer0.96
    4. EUROPE1.00
    5. EUROPE1.00
    6. Unblock1.00
    7. AlEngineer0.96
    8. SAFE1.00
    9. AlEngineer0.97
    10. EUROPE1.00
    11. INTELLIGENCE1.00
    12. EUROPE1.00
    13. Google DeepMind1.00
    14. Sonc1.00
    15. Engineer1.00
    16. EUROPE1.00
    17. AlEngineer0.98
    18. EUROPE1.00
    19. ai0.71
    20. PRESENTED BY1.00
    21. AlEngineer0.97
    22. ight data0.97
    23. AlEngineer0.99
    24. Google DeepMind1.00
    25. EUROPE1.00
    26. EUROPE1.00
    27. Snorke!1.00
    28. AlEngineer0.99
    29. eNCORD0.97
    30. EUROPE0.99

  • 0:16 #4 done31 line(s)

    shot 4·sharpness 875.8

    1. AIEnginee0.94
    2. OpenAl0.95
    3. AlEngineer0.96
    4. EUROPE0.89
    5. Unbl-0.99
    6. AlEngineer0.95
    7. AFE1.00
    8. AlEngineer0.98
    9. Op1.00
    10. BUMOPE0.89
    11. ITELLIGENCE0.97
    12. EUROPE0.91
    13. Sonar1.00
    14. gineer1.00
    15. Google DeepMind1.00
    16. ANE0.81
    17. AlEngineer0.99
    18. EUROPE1.00
    19. PRESENTED BY0.97
    20. AlEngineer0.98
    21. ;htdata0.98
    22. AIEngineer0.96
    23. Bra0.97
    24. Google DeepMind0.99
    25. BUROPE0.94
    26. EUROPE0.82
    27. Snorkel0.97
    28. gineer1.00
    29. eNCORD0.96
    30. ANE0.79
    31. YOFE0.84

  • 0:18 #5 done33 line(s)

    shot 5·sharpness 980.5

    1. OpenAI0.94
    2. AlEngineer0.99
    3. EUROPE1.00
    4. AlEngineer0.98
    5. SAFE1.00
    6. -ineer0.93
    7. OpenAI0.87
    8. AIEngin0.96
    9. EUROPE1.00
    10. INTELLIGENCE1.00
    11. EUROPE1.00
    12. Sonar1.00
    13. AlEngineer0.98
    14. Mind1.00
    15. AlEngineer0.99
    16. Wor1.00
    17. EUROPE1.00
    18. EUROPE1.00
    19. AlEngineer0.97
    20. bright data1.00
    21. gineer1.00
    22. Braintrust1.00
    23. AIEngin0.95
    24. EUROPE1.00
    25. ROPE1.00
    26. EUROPE1.00
    27. Snorkel0.99
    28. AlEngineer1.00
    29. NCORD1.00
    30. AlEngineer0.98
    31. tails1.00
    32. EUROPE1.00
    33. EUROPE1.00

  • 0:22 #6 done22 line(s)

    shot 6·sharpness 2145.1

    1. nAl0.93
    2. AlEngineer0.99
    3. EUROPE1.00
    4. ineer0.98
    5. SAFE1.00
    6. PE1.00
    7. INTELLIGENCE1.00
    8. Engineer1.00
    9. EUROPE1.00
    10. A Piece Of PI0.97
    11. ineer0.99
    12. ight data1.00
    13. Embedding The OpenClaw Coding Agent In Your Product0.99
    14. PE1.00
    15. orkel1.00
    16. AlEngineer0.97
    17. EUROPE1.00
    18. TAVON.ai1.00
    19. [email protected]· x.com/luebken0.98
    20. Google DeepMind0.98
    21. AlEngineer0.98
    22. EUROPE1.00

  • 0:37 #7 done16 line(s)

    shot 7·sharpness 2100.7

    1. therai1.00
    2. Engineer1.00
    3. BUROPE0.94
    4. Nork0.99
    5. A Piece Of PI0.95
    6. Embedding The OpenClaw Coding Agent In Your Product1.00
    7. ngineer1.00
    8. EUROPE0.99
    9. ze1.00
    10. ENTED BY0.99
    11. DeepMind0.99
    12. TAVON.ai1.00
    13. [email protected]· x.com/luebken0.98
    14. Google DeepMind0.99
    15. AlEngineer0.99
    16. EUROPE1.00

  • 0:47 #8 done15 line(s)

    shot 8·sharpness 1916.0

    1. OpenA'0.91
    2. AlEngineer0.99
    3. EUROPE1.00
    4. AlEngir0.95
    5. WorkO0.99
    6. EUROP1.00
    7. AlEngineer0.94
    8. EUROPE1.00
    9. PRESENTED BY1.00
    10. Google DeepMir1.00
    11. AlEngineer0.98
    12. # Braintrust0.98
    13. WorkOS1.00
    14. OpenAl0.95
    15. EUROPE1.00

  • 1:03 #9 done57 line(s)

    shot 9·sharpness 1427.4

    1. AlEngineer0.99
    2. EUROPE1.00
    3. TAVON.ai1.00
    4. AlEngineer0.97
    5. togetherai0.99
    6. Put Al Agents to Work0.98
    7. EUROPE1.00
    8. Worl0.99
    9. ..gineer0.92
    10. EUROPE1.00
    11. Consulting1.00
    12. Implementation1.00
    13. Toolkit1.00
    14. IEngineer0.96
    15. urize1.00
    16. Find the right use cases and build a0.98
    17. strategy1.00
    18. We build, deploy, and optimize1.00
    19. agents for you1.00
    20. Everything you need to build0.99
    21. powerful agents1.00
    22. EUROPE1.00
    23. Not sure where to start? We help you1.00
    24. Need a complete solution? We design,0.99
    25. We provide the frameworks,0.99
    26. identify high-impact opportunities,1.00
    27. develop, and integrate custom agents0.99
    28. templates, and building blocks to0.99
    29. RESENTED BY1.00
    30. run controlled experiments, and build0.99
    31. a roadmap for Al agent adoption.1.00
    32. tailored to your workflows and0.99
    33. systems.1.00
    34. create production-ready Al agents —0.98
    35. so your team can move from idea to1.00
    36. gle DeepMind1.00
    37. deployment fast.0.99
    38. Includes:1.00
    39. Includes:1.00
    40. Includes:1.00
    41. AlEngineer0.99
    42. Discovery workshops1.00
    43. Custom agent development1.00
    44. Agent development frameworks1.00
    45. EUROPE1.00
    46. ✓Use case identification0.97
    47. System integration1.00
    48. Pre-built templates & components0.99
    49. Pilot programs (4-week sprints)1.00
    50. Workflow automation0.98
    51. Integration libraries0.99
    52. Strategic advisory1.00
    53. √Ongoing optimization0.97
    54. √Best practices & documentation0.98
    55. Engineering the future of Al1.00
    56. AlEngineer0.97
    57. EUROPE1.00

  • 1:09 #10 done14 line(s)

    shot 10·sharpness 1337.8

    1. Google DeepMind1.00
    2. AlEngineer0.94
    3. AlEnni0.79
    4. Trigger.d1.00
    5. AlEngineer0.96
    6. come1.00
    7. AK0.77
    8. E0.85
    9. AlEngineer0.99
    10. neo0.83
    11. Ken Thompson1.00
    12. Engineering the future of Al1.00
    13. AlEngineer0.99
    14. EUROPE1.00

  • 1:20 #11 done14 line(s)

    shot 11·sharpness 1502.4

    1. BARERDIL0.87
    2. toge1.00
    3. We are in the fuck1.00
    4. around and find out1.00
    5. phase of (coding) agents1.00
    6. AlEngineer0.99
    7. THESIS 11.00
    8. EUROPE1.00
    9. Λar0.94
    10. PRESENTED BY1.00
    11. Google DeepMind0.98
    12. Google DeepMind1.00
    13. AlEngineer0.99
    14. EUROPE1.00

  • 1:24 #12 done11 line(s)

    shot 12·sharpness 1409.0

    1. AlEngh0.85
    2. Unbl0.99
    3. ANEngh0.78
    4. We are in the fuck1.00
    5. around and find out0.99
    6. phase of (coding) agents0.98
    7. AlEngh0.86
    8. THESIS 11.00
    9. Google DeepMind0.99
    10. AlEngineer0.99
    11. EUROPE1.00

  • 1:42 #13 done23 line(s)

    shot 13·sharpness 1523.7

    1. nAl0.78
    2. AlEngineer0.97
    3. EUROPE1.00
    4. neer1.00
    5. SAFE1.00
    6. BARENDIL0.87
    7. INTELLIGENCE1.00
    8. ngineer1.00
    9. EUROPE1.00
    10. We are in the fuck1.00
    11. around and find out0.99
    12. neer1.00
    13. ight data1.00
    14. phase of (coding) agents1.00
    15. THESIS 11.00
    16. orkel0.97
    17. Engineer1.00
    18. EUROPE1.00
    19. AlEngineer0.98
    20. # Braintrust0.97
    21. WorkOS1.00
    22. OpenAl0.94
    23. EUROPE1.00

  • 2:04 #14 done20 line(s)

    shot 14·sharpness 1911.0

    1. neer0.99
    2. OpenAI0.92
    3. locked0.97
    4. AlEngineer0.96
    5. EUROPE1.00
    6. Sonar1.00
    7. Engineer1.00
    8. Unix philosophy1.00
    9. EUROPE1.00
    10. RESENTED BY1.00
    11. ngineer0.98
    12. gle DeepMind0.97
    13. UROPE1.00
    14. "Write programs that do one0.98
    15. thing and do it well."1.00
    16. Snorkel0.95
    17. Ken Thompson1.00
    18. Engineering the future of Al1.00
    19. AlEngineer0.97
    20. EUROPE1.00

  • 2:22 #15 done44 line(s)

    shot 15·sharpness 1622.0

    1. WorkOS1.00
    2. Chat1.00
    3. Cowork1.00
    4. Code1.00
    5. 日0.56
    6. Demo Project / Build Q1 2026 sensitivity analysis matrix0.99
    7. Q0.97
    8. Progress1.00
    9. and-design-sensitivitymatrix0.96
    10. Anaiyze-existing forecast data0.98
    11. 色0.60
    12. Build a sensitivity analysis matrix for Forecast_Q1_2026_Weekly_Update1.00
    13. Building sensitivity analysis0.96
    14. spreadsheet1.00
    15. Recalculate formulas and1.00
    16. 白0.91
    17. spreadsheet structure and contents1.00
    18. verify for errors1.00
    19. Read Forecast_Q1_2026_Weekly_Update.xlsx contents0.98
    20. Verify output and deliver file0.99
    21. Script0.99
    22. eer1.00
    23. E Update todo list0.96
    24. demo1.00
    25. □0.73
    26. Rosult0.91
    27. Instructions ·CLAUDE.md0.98
    28. pMind1.00
    29. BY1.00
    30. Running command0.99
    31. Context1.00
    32. Skills0.95
    33. xlsx0.95
    34. Skills1.00
    35. lodal1.00
    36. xlsx0.99
    37. Reply...0.97
    38. +0.77
    39. Opus 4.61.00
    40. Claude is Al and can make mistakes. Please double-check responses.0.98
    41. https://github.com/anthropics/skills/tree/main/skills/xlsx1.00
    42. Google DeepMind1.00
    43. AlEngineer0.96
    44. EUROPE1.00

  • 3:08 #16 done44 line(s)

    shot 16·sharpness 1788.9

    1. □0.80
    2. Chat1.00
    3. Cowork1.00
    4. Code1.00
    5. □0.63
    6. Demo Project / Build Q1 2026 sensitivity analysis matrix0.98
    7. Q0.95
    8. Progress1.00
    9. Analyze existing forecast data0.96
    10. and-design-sensitivity matrix0.96
    11. pandas1.00
    12. 回0.56
    13. Build a sensitivity analysis matrix for Forecast_Q1_2026_Weekly_Update0.99
    14. Building sensitivity analysis1.00
    15. spreadsheet1.00
    16. ②0.73
    17. Recalculate formulas and0.99
    18. è0.83
    19. Analyzed Q1 2026 forecast spreadsheet structure and contents0.99
    20. verify for errors0.99
    21. openpyxl1.00
    22. Read Forecast_Q1_2026_Weekly_Update.xlsx contents0.99
    23. Verify output and deliver file1.00
    24. Script1.00
    25. demo1.00
    26. □0.67
    27. LibreOffice1.00
    28. Update todo list0.97
    29. Result0.99
    30. Instructions·CLAUDE.md1.00
    31. The Document Foundation0.99
    32. Running command1.00
    33. Context1.00
    34. Skills1.00
    35. Skills1.00
    36. xlsx0.97
    37. xlsx0.98
    38. Reply...0.99
    39. +0.87
    40. Opus 4.61.00
    41. L Queue0.91
    42. M1.00
    43. Claude is Al and can make mistakes. Please double-check responses.1.00
    44. https://github.com/anthropics/skills/tree/main/skills/xlsx1.00

  • 3:34 #17 done23 line(s)

    shot 17·sharpness 1925.0

    1. ineer0.99
    2. WorkOS1.00
    3. Your agent will hallucinate these patterns anyway0.99
    4. FE1.00
    5. AlEngineer0.98
    6. LIGENCE1.00
    7. EUROPE1.00
    8. arize1.00
    9. AlEngineer1.00
    10. EUROPE1.00
    11. PRESENTED BY1.00
    12. Engineer1.00
    13. ogle DeepMind1.00
    14. EUROPE1.00
    15. Enterprise Coding1.00
    16. Modal1.00
    17. Agent Patterns1.00
    18. Authoritative Advice1.00
    19. O RLY?0.96
    20. Foreword by an Al0.96
    21. Google DeepMind1.00
    22. AlEngineer0.99
    23. EUROPE1.00

  • 3:48 #18 done19 line(s)

    shot 18·sharpness 1650.4

    1. kOS0.84
    2. AlEngineer0.97
    3. EUROPE1.00
    4. eer1.00
    5. Braintrust1.00
    6. IEngineer0.98
    7. EUROPE1.00
    8. First Architectural Pattern:1.00
    9. neer0.96
    10. igger.dev1.00
    11. Make It Easy for Coding Agents.1.00
    12. lodal0.91
    13. Engineer1.00
    14. EUROPE1.00
    15. AlEngineer0.97
    16. # Braintrust0.96
    17. WorkOS1.00
    18. OpenAl0.96
    19. EUROPE1.00

  • 4:06 #19 done23 line(s)

    shot 19·sharpness 1848.1

    1. Engineer1.00
    2. Google Deep1.00
    3. EUROPE1.00
    4. Agenda1.00
    5. unpor0.92
    6. AlEngineer0.95
    7. EUROPE1.00
    8. rigger.o0.98
    9. AlEngineer0.99
    10. 1. Intro Pl0.97
    11. 4. Pl in OpenClaw0.96
    12. EUROPE1.00
    13. Google DeepMind0.98
    14. PRESENTED BY1.00
    15. AlEngineer0.97
    16. 2. Agents0.99
    17. 5. Pl in My App0.95
    18. EUROPE1.00
    19. 3. Coding Agents0.99
    20. CLOUDFI0.98
    21. AlEngineer0.96
    22. Engineering the future of Al1.00
    23. EUROPE1.00

  • 4:19 #20 done40 line(s)

    shot 20·sharpness 3777.7

    1. gineer1.00
    2. Google DeepM0.93
    3. JROPE0.99
    4. pi.dev1.00
    5. uopor0.76
    6. AlEngineer0.97
    7. EUROPE1.00
    8. Pi is a minimal terminal coding0.97
    9. F0.82
    10. harness.[...] skips features like sub0.99
    11. Trigger.de0.96
    12. agents and plan mode. Instead, you1.00
    13. AlEngineer0.99
    14. There are many coding agents, but this one1.00
    15. can ask pi to build what you want0.97
    16. EUROPE1.00
    17. is mine.1.00
    18. or install a third party pi package [...]0.98
    19. PRESENTED BY1.00
    20. AlEngineer0.97
    21. GitHub1.00
    22. npmDiscord0.99
    23. Google DeepMind1.00
    24. EUROPE1.00
    25. $ npm install -g @mariozechner/pi-coding-agent0.98
    26. Copy1.00
    27. CLOUDFL1.00
    28. • Open Source (MIT)1.00
    29. Read & Understand Code - I can explore any codebase, understand architecture, find bugs0.99
    30. Execute Commands - Run tests, builds, git operations, any shell command0.98
    31. • Build by Mario Zechner0.99
    32. Edit Files0.99
    33. - Make precise surgical edits or write new files0.98
    34. Find files, grep for patterns, trace dependencies1.00
    35. • As of this week part of earendil.com0.99
    36. Skills0.97
    37. Google Calendar/Drive/Gmail, transcription, and more0.99
    38. Google DeepMind1.00
    39. AlEngineer0.98
    40. EUROPE1.00

  • 4:44 #21 done24 line(s)

    shot 21·sharpness 3922.3

    1. pi.dev1.00
    2. Pi is a minimal terminal coding0.98
    3. P0.95
    4. harness.[...] skips features like sub0.98
    5. agents and plan mode. Instead, you1.00
    6. There are many coding agents, but this one1.00
    7. can ask pi to build what you want0.99
    8. is mine.1.00
    9. or install a third party pi package [...]0.98
    10. GitHub1.00
    11. npm1.00
    12. Discord1.00
    13. $ npm install -g @mariozechner/pi-coding-agent0.99
    14. Copy1.00
    15. Mario Rodler1.00
    16. Here's what I can do:1.00
    17. • Open Source (MIT)1.00
    18. Read & Understand Code - I can explore any codebase, understand architecture, find bugs0.99
    19. Commands - Run tests, builds, git operations, any shell command0.99
    20. Build by Mario Zechner0.98
    21. Edit Files - Make precise surgical edits or write new files1.00
    22. Search & Navigate - Find files, grep for patterns, trace dependencies0.99
    23. • As of this week part of earendil.com0.99
    24. skills - Web search, browser automation, Google Calendar/Drive/Gmail, transcription, and more0.99

  • 5:06 #22 done84 line(s)

    shot 22·sharpness 1255.9

    1. Packages - pi.dev0.99
    2. x0.82
    3. +0.87
    4. keywords:pi-package - npm: x0.95
    5. +0.98
    6. ←1.00
    7. →0.99
    8. ×0.79
    9. pi.dev/packages0.97
    10. ☆1.00
    11. Work1.00
    12. ←1.00
    13. →0.89
    14. npmjs.com/search?q=keywords:pi-package1.00
    15. ☆0.97
    16. Work1.00
    17. Pro1.00
    18. Teams1.00
    19. Pricing1.00
    20. Documentation1.00
    21. P0.75
    22. npm1.00
    23. Q keywords:pi-package0.97
    24. Search1.00
    25. Sign Up0.99
    26. Sign In0.99
    27. 1000+ packages found1.00
    28. Sort by: Default1.00
    29. GitHub1.00
    30. npm1.00
    31. Discord1.00
    32. taskplane1.00
    33. Al agent orchestration for pi — parallel task execution with checkpoint discipline0.98
    34. Packages1.00
    35. pi-package1.00
    36. agent1.00
    37. orchestration1.00
    38. task-runner1.00
    39. parallel1.00
    40. 17,0351.00
    41. Extensions, skills, themes, and more from the community.0.99
    42. henrylach • 0.24.30• 15 hours ago· 0 dependents • && MIT0.90
    43. Search packages...1.00
    44. Most downloads1.00
    45. @codexstar/pi-pompom1.00
    46. extension1.00
    47. skill1.00
    48. thene1.00
    49. prompt1.00
    50. has demo0.96
    51. 1085 of 10851.00
    52. Pi-Pompom — A 3D raymarched virtual pet with voice, ambient weather sounds, Al side chat,0.99
    53. agent tracking, and 23 layered sound effects for Pi CLI1.00
    54. RECENT1.00
    55. pi-package1.00
    56. pi-extension1.00
    57. pompom1.00
    58. companion1.00
    59. virtual-pet1.00
    60. 19,2161.00
    61. @dreki-gg/pi-modes1.00
    62. pi-yank1.00
    63. @feniix/pi-notion1.00
    64. terminal-pet1.00
    65. terminal-art1.00
    66. raymarching1.00
    67. ascii-art1.00
    68. ansi0.99
    69. tui1.00
    70. widget1.00
    71. Config-driven preset and mode s..0.97
    72. Lightweight /yank extension for p..1.00
    73. Notion API extension for pi0.98
    74. interactive1.00
    75. View more1.00
    76. 21m ago0.96
    77. 38n ago0.97
    78. 42m ago0.98
    79. extension0.97
    80. 商0.54
    81. engaze • 7.8.36· 2 days ago · 0 dependents · & MIT0.90
    82. @tryinget/pi-interaction-kit1.00
    83. pi-subagents1.00
    84. Pi extension for delegating tasks to subagents with chains, parallel execution, and TUI1.00

  • 5:31 #23 done6 line(s)

    shot 23·sharpness 1539.4

    1. eer1.00
    2. Google DeepMing0.96
    3. pod1.00
    4. AlEngineer0.98
    5. EUROPE1.00
    6. igger.dev1.00

The page's on-screen-text budget of 600 lines is spent, so the last cards in this grid list fewer lines than they hold. Narrow the page with ?frames= to read them.

Transcript

235 cues· 2,950 words· 15,659 chars

  1. 0:15 All right.
  2. 0:16 I was introduced to Py by looking into OpenClaw.
  3. 0:21 There was a conference, a meetup, and I said, OK, we're doing OpenClaw.
  4. 0:26 And I wasn't so much interested into all the craziness things that people are doing, but I was more interested in understanding of how these things work.
  5. 0:36 So I was looking into Py and understand the whole world of what Py is able to do.
  6. 0:44 This is the one picture you need to take.
  7. 0:46 Please feel free to take more pictures, but all the slides and the examples are there.
  8. 0:52 So that's the one slide.
  9. 0:54 All right, very quick about myself.
  10. 0:58 We're creating a small company, Tavon AI.
  11. 1:00 We're building agents for organizations, small out of Europe, but getting started.
  12. 1:06 And what I really like about, sorry,
  13. 1:13 What I really like about Mario's talk is this quote.
  14. 1:19 You probably have seen this this morning.
  15. 1:21 We are in the fuck around and find our own face for coding agents.
  16. 1:27 So everything that I'm going to show you is what I know today.
  17. 1:32 And I'm going to do the talk again in a couple of weeks.
  18. 1:35 And it's going to most likely be different.
  19. 1:38 But as Mario was showing this morning, he has created this minimal set, this coding agent that is available for you guys to fool around with.
  20. 1:52 And that's what I'd like to encourage you.
  21. 1:55 So coding agents, and why is it so exciting for us to build more products?
  22. 2:00 This is Ken Thompson, inventor of Unix.
  23. 2:04 And this is the famous quote by him, one of the quotes, write programs that do one thing and one thing well.
  24. 2:12 And I really like that because that kind of works to our advantage with agents.
  25. 2:17 And the best part where I show this is with cowork.
  26. 2:21 So this is cowork, Claude's desktop.
  27. 2:25 And they're basically bundling their coding agent into something where they feel is more applicable.
  28. 2:31 And to be honest, I've seen very good receptions around this.
  29. 2:35 And when you use it with financing tools, with their finance tools, you always need to work with Excel, right?
  30. 2:42 So they have this Excel skill down there, and it talks to Excel, right?
  31. 2:49 Well, it doesn't.
  32. 2:51 Instead, it uses a set of small tools, small CLIs, pandas, OpenPyXL, stuff from LibreOffice, and package this into their own skill to make it up and running.
  33. 3:05 And I think this is a great example to kind of get your thoughts going of what is doable.
  34. 3:14 I haven't written a book, and nobody can write a book about this, right?
  35. 3:18 Because there are no patterns, right?
  36. 3:20 We need to figure this out.
  37. 3:22 We're seeing some emerging patterns in the coding space, right?
  38. 3:25 There's obviously tons of different coding agents, and we're seeing this.
  39. 3:29 But there is no authoritative resource around this, right?
  40. 3:32 So get going.
  41. 3:34 One thing when I was talking to Ivan yesterday we realized is like one architectural pattern that we're seeing is that make it easy for coding agents right now that is very broad but think about it right like make not don't try to be you know very complex and things but think about.
  42. 3:55 The coding agent, what is it good at?
  43. 3:58 And how do I build my system so that the agent is easy, make it accessible?
  44. 4:03 And I have some examples.
  45. 4:05 All right, this is the rough agenda.
  46. 4:08 for the next 10 minutes or so.
  47. 4:11 I'm not gonna talk too much about Py in OpenClaw.
  48. 4:14 I have two slides, slides are online, so we'll take it from there.
  49. 4:17 So again, very brief introduction of Py.
  50. 4:21 Mario, great work.

Chapters

  1. 0:00 <Untitled Chapter 1>
  2. 0:15 Introduction to Pi and OpenClaw
  3. 1:55 The philosophy of coding agents (doing one thing well)
  4. 3:34 Architectural pattern: Making systems easy for agents
  5. 5:13 Defining an agent: LLM with tools in a loop
  6. 6:37 Practical example: CRM lead qualifier
  7. 8:41 Coding agents vs. core agents
  8. 10:06 Extension API and UI interactions
  9. 12:53 Multi-channel environment: Pi and OpenClaw
  10. 14:46 Real-world B2B sales pipeline application
  11. 18:14 Demonstration: Dashboard and email drafting process
  12. 20:00 Final takeaways and encouragement to tinker

Open at this second