vidtheque.

read-only demo

Videos LZuWZRze3MU

The New Primitives: Building AI Native Software — Kwindla Kramer, Daily

index_state ready data_status ok

AI Engineer· published 2026-08-07· 0:21:14· en-US· indexed 2026-08-10 19:34

Open on YouTube

Scene timeline

  1. Shot 0, 0:00 to 0:03, 1 of 1 keyframes kept
  2. Shot 1, 0:03 to 0:05, 1 of 1 keyframes kept
  3. Shot 2, 0:05 to 0:12, 1 of 1 keyframes kept
  4. Shot 3, 0:12 to 0:20, 1 of 1 keyframes kept
  5. Shot 4, 0:20 to 1:08, 1 of 1 keyframes kept
  6. Shot 5, 1:08 to 1:32, 1 of 1 keyframes kept
  7. Shot 6, 1:32 to 2:12, 1 of 1 keyframes kept
  8. Shot 7, 2:12 to 2:27, 0 of 1 keyframes kept
  9. Shot 8, 2:27 to 2:40, 1 of 1 keyframes kept
  10. Shot 9, 2:40 to 3:08, 1 of 1 keyframes kept
  11. Shot 10, 3:08 to 3:36, 1 of 1 keyframes kept
  12. Shot 11, 3:36 to 3:53, 1 of 1 keyframes kept
  13. Shot 12, 3:53 to 4:33, 1 of 1 keyframes kept
  14. Shot 13, 4:33 to 4:56, 1 of 1 keyframes kept
  15. Shot 14, 4:56 to 5:25, 1 of 1 keyframes kept
  16. Shot 15, 5:25 to 5:55, 0 of 1 keyframes kept
  17. Shot 16, 5:55 to 6:18, 0 of 1 keyframes kept
  18. Shot 17, 6:18 to 6:43, 0 of 1 keyframes kept
  19. Shot 18, 6:43 to 7:08, 0 of 1 keyframes kept
  20. Shot 19, 7:08 to 7:29, 1 of 1 keyframes kept
  21. Shot 20, 7:29 to 7:35, 1 of 1 keyframes kept
  22. Shot 21, 7:35 to 7:48, 1 of 1 keyframes kept
  23. Shot 22, 7:48 to 7:49, 1 of 1 keyframes kept
  24. Shot 23, 7:49 to 7:57, 1 of 1 keyframes kept
  25. Shot 24, 7:57 to 7:59, 1 of 1 keyframes kept
  26. Shot 25, 7:59 to 8:28, 0 of 1 keyframes kept
  27. Shot 26, 8:28 to 8:58, 1 of 1 keyframes kept
  28. Shot 27, 8:58 to 9:48, 1 of 1 keyframes kept
  29. Shot 28, 9:48 to 9:53, 1 of 1 keyframes kept
  30. Shot 29, 9:53 to 10:40, 1 of 1 keyframes kept
  31. Shot 30, 10:40 to 10:56, 1 of 1 keyframes kept
  32. Shot 31, 10:56 to 11:18, 1 of 1 keyframes kept
  33. Shot 32, 11:18 to 11:21, 1 of 1 keyframes kept
  34. Shot 33, 11:21 to 11:26, 1 of 1 keyframes kept
  35. Shot 34, 11:26 to 11:30, 1 of 1 keyframes kept
  36. Shot 35, 11:30 to 11:36, 1 of 1 keyframes kept
  37. Shot 36, 11:36 to 11:40, 1 of 1 keyframes kept
  38. Shot 37, 11:40 to 12:24, 0 of 1 keyframes kept
  39. Shot 38, 12:24 to 12:52, 0 of 1 keyframes kept
  40. Shot 39, 12:52 to 13:21, 0 of 1 keyframes kept
  41. Shot 40, 13:21 to 14:03, 1 of 1 keyframes kept
  42. Shot 41, 14:03 to 14:32, 1 of 1 keyframes kept
  43. Shot 42, 14:32 to 15:06, 1 of 1 keyframes kept
  44. Shot 43, 15:06 to 15:09, 1 of 1 keyframes kept
  45. Shot 44, 15:09 to 15:16, 1 of 1 keyframes kept
  46. Shot 45, 15:16 to 15:20, 1 of 1 keyframes kept
  47. Shot 46, 15:20 to 15:26, 1 of 1 keyframes kept
  48. Shot 47, 15:26 to 15:28, 0 of 1 keyframes kept
  49. Shot 48, 15:28 to 15:32, 1 of 1 keyframes kept
  50. Shot 49, 15:32 to 15:36, 1 of 1 keyframes kept
  51. Shot 50, 15:36 to 15:48, 1 of 1 keyframes kept
  52. Shot 51, 15:48 to 15:50, 0 of 1 keyframes kept
  53. Shot 52, 15:50 to 15:51, 0 of 1 keyframes kept
  54. Shot 53, 15:51 to 15:53, 0 of 1 keyframes kept
  55. Shot 54, 15:53 to 15:56, 0 of 1 keyframes kept
  56. Shot 55, 15:56 to 15:58, 1 of 1 keyframes kept
  57. Shot 56, 15:58 to 16:01, 1 of 1 keyframes kept
  58. Shot 57, 16:01 to 16:05, 1 of 1 keyframes kept
  59. Shot 58, 16:05 to 16:07, 0 of 1 keyframes kept
  60. Shot 59, 16:07 to 16:09, 0 of 1 keyframes kept
  61. Shot 60, 16:09 to 16:13, 1 of 1 keyframes kept
  62. Shot 61, 16:13 to 16:17, 1 of 1 keyframes kept
  63. Shot 62, 16:17 to 16:19, 0 of 1 keyframes kept
  64. Shot 63, 16:19 to 16:24, 1 of 1 keyframes kept
  65. Shot 64, 16:24 to 16:26, 1 of 1 keyframes kept
  66. Shot 65, 16:26 to 16:28, 1 of 1 keyframes kept
  67. Shot 66, 16:28 to 16:29, 1 of 1 keyframes kept
  68. Shot 67, 16:29 to 16:34, 1 of 1 keyframes kept
  69. Shot 68, 16:34 to 16:36, 1 of 1 keyframes kept
  70. Shot 69, 16:36 to 16:38, 1 of 1 keyframes kept
  71. Shot 70, 16:38 to 16:41, 1 of 1 keyframes kept
  72. Shot 71, 16:41 to 16:43, 1 of 1 keyframes kept
  73. Shot 72, 16:43 to 16:47, 1 of 1 keyframes kept
  74. Shot 73, 16:47 to 16:49, 1 of 1 keyframes kept
  75. Shot 74, 16:49 to 16:51, 1 of 1 keyframes kept
  76. Shot 75, 16:51 to 16:55, 1 of 1 keyframes kept
  77. Shot 76, 16:55 to 17:01, 1 of 1 keyframes kept
  78. Shot 77, 17:01 to 17:08, 1 of 1 keyframes kept
  79. Shot 78, 17:08 to 17:10, 1 of 1 keyframes kept
  80. Shot 79, 17:10 to 17:29, 0 of 1 keyframes kept
  81. Shot 80, 17:29 to 17:37, 0 of 1 keyframes kept
  82. Shot 81, 17:37 to 18:08, 0 of 1 keyframes kept
  83. Shot 82, 18:08 to 18:22, 1 of 1 keyframes kept
  84. Shot 83, 18:22 to 18:26, 1 of 1 keyframes kept
  85. Shot 84, 18:26 to 18:29, 1 of 1 keyframes kept
  86. Shot 85, 18:29 to 18:55, 1 of 1 keyframes kept
  87. Shot 86, 18:55 to 19:31, 0 of 1 keyframes kept
  88. Shot 87, 19:31 to 19:33, 0 of 1 keyframes kept
  89. Shot 88, 19:33 to 19:36, 0 of 1 keyframes kept
  90. Shot 89, 19:36 to 19:39, 0 of 1 keyframes kept
  91. Shot 90, 19:39 to 19:41, 0 of 1 keyframes kept
  92. Shot 91, 19:41 to 19:44, 0 of 1 keyframes kept
  93. Shot 92, 19:44 to 19:46, 1 of 1 keyframes kept
  94. Shot 93, 19:46 to 19:55, 0 of 1 keyframes kept
  95. Shot 94, 19:55 to 20:06, 0 of 1 keyframes kept
  96. Shot 95, 20:06 to 20:12, 0 of 1 keyframes kept
  97. Shot 96, 20:12 to 20:17, 0 of 1 keyframes kept
  98. Shot 97, 20:17 to 20:20, 0 of 1 keyframes kept
  99. Shot 98, 20:20 to 20:29, 0 of 1 keyframes kept
  100. Shot 99, 20:29 to 20:49, 1 of 1 keyframes kept
  101. Shot 100, 20:49 to 20:52, 0 of 1 keyframes kept
  102. Shot 101, 20:52 to 20:57, 1 of 1 keyframes kept
  103. Shot 102, 20:57 to 21:13, 0 of 1 keyframes kept

103 shot(s).

keyframes kept every frame deduplicated

What was stored

cues
205
whisperx 205
chunks
36
from 205 cues
keyframes
69
kept of 103 captured
frames with text
69
810 lines read
chapters
16
from the source metadata
keyframe bytes
12.8 MB
word timings on 205 cues

Provenance

Each pipeline stage, its state and the model that produced it
stage state model started took
fetch done — 2026-08-09 21:08 0s
stt done — 2026-08-08 23:33 23s
chunk done — 2026-08-08 23:33 0s
text_embed done — 2026-08-10 19:34 1s
keyframe done — 2026-08-08 23:33 3m 28s
ocr done — 2026-08-08 23:36 20s
frame_embed done — 2026-08-10 19:34 11s

Frames, and what the machine read

  • 0:02 #0 done2 line(s)

    shot 0·sharpness 454.1

    1. AlEngineer0.95
    2. World's Fair0.97

  • 0:03 #1 done2 line(s)

    shot 1·sharpness 665.2

    1. AIEngineer0.95
    2. World's Fair0.99

  • 0:10 #2 done24 line(s)

    shot 2·sharpness 2746.2

    1. LAB & PLATINUM SPONSORS0.99
    2. Amazon AGI Lab0.98
    3. ANTHROP\C1.00
    4. Google DeepMind1.00
    5. MINIMAX0.94
    6. OpenAI0.92
    7. Akamai1.00
    8. arize0.93
    9. aws1.00
    10. Braintrust bright data0.98
    11. B1.00
    12. Browserbase1.00
    13. docker1.00
    14. :neo4j0.92
    15. ORACLE1.00
    16. PayPal1.00
    17. qodo1.00
    18. reducto1.00
    19. Sonar1.00
    20. Makers of0.99
    21. togetherai1.00
    22. Unblocked1.00
    23. WorkOS1.00
    24. SonarQube1.00

  • 0:17 #3 done2 line(s)

    shot 3·sharpness 168.8

    1. AlEngineer0.99
    2. World's Fair0.98

  • 1:02 #4 done13 line(s)

    shot 4·sharpness 1876.3

    1. AlEngineer0.99
    2. World'sFair0.96
    3. PRESENTED BY1.00
    4. The New0.97
    5. Microsoft1.00
    6. Primitives1.00
    7. Building Al-Native Software1.00
    8. Kwindla Hultman Kramer1.00
    9. @kwindla1.00
    10. Daily.co |Pipecat0.99
    11. TRACK 6· JUNE 30, 20260.96
    12. World'sFair1.00
    13. Voice & Realtime Al0.99

  • 1:15 #5 done12 line(s)

    shot 5·sharpness 1860.0

    1. AlEngineer0.99
    2. 19451.00
    3. 021.00
    4. World'sFair1.00
    5. PRESENTED BY1.00
    6. Vannevar Bush1.00
    7. Microsoft1.00
    8. As We May Think0.97
    9. GQ0.71
    10. TRACK 6· JUNE 30, 20260.96
    11. World'sFair1.00
    12. Voice & Realtime Al0.98

  • 1:48 #6 done23 line(s)

    shot 6·sharpness 2506.3

    1. AlEngineer0.98
    2. AS WE MAY THINK0.99
    3. 031.00
    4. World's Fair0.96
    5. Document display on a screen0.98
    6. Data transmission over networks, including by0.99
    7. radio1.00
    8. Document scanning and optical character0.99
    9. recognition1.00
    10. The personal computer0.99
    11. • Speech to text0.97
    12. GoPro camera0.99
    13. Text to speech1.00
    14. The Amazon Kindle store0.99
    15. • Programming languages0.98
    16. • Voice interfaces0.96
    17. • Hypertext0.94
    18. Brain-computer interfaces0.98
    19. • Search engines0.97
    20. GノQ0.58
    21. TRACK 6· JUNE 30, 20260.97
    22. World's Fair0.98
    23. Voice & Realtime Al0.99

  • 2:24 #7 skipped

    shot 7·duplicate of #6

  • 2:36 #8 done14 line(s)

    shot 8·sharpness 1716.8

    1. AlEngineer0.99
    2. 061.00
    3. World'sFair1.00
    4. Loops1.00
    5. PRIORS0.88
    6. Tool calls1.00
    7. Context engineering1.00
    8. LLMs1.00
    9. GPUs1.00
    10. NO-PRIORS.COM1.00
    11. GノQ0.59
    12. TRACK 6· JUNE 30, 20260.96
    13. World'sFair1.00
    14. Voice & Realtime Al0.98

  • 2:59 #9 done12 line(s)

    shot 9·sharpness 1654.0

    1. AlEngineer0.99
    2. World'sFair1.00
    3. Loops1.00
    4. PRIORS0.86
    5. Tool calls1.00
    6. Context engineering0.99
    7. LLMs1.00
    8. GPUs1.00
    9. NO-PRIORS.COM1.00
    10. TRACK 6· JUNE 30, 20260.96
    11. World'sFair0.97
    12. Voice & Realtime Al0.99

  • 3:14 #10 done12 line(s)

    shot 10·sharpness 1640.8

    1. AlEngineer0.99
    2. World'sFair1.00
    3. Loops1.00
    4. PRIORS0.89
    5. Tool calls1.00
    6. Context engineering1.00
    7. LLMs1.00
    8. GPUs1.00
    9. NO-PRIORS.COM1.00
    10. TRACK 6· JUNE 30, 20260.96
    11. World'sFair1.00
    12. Voice & Realtime Al0.99

  • 3:51 #11 done10 line(s)

    shot 11·sharpness 1765.2

    1. AlEngineer0.97
    2. Agents ++1.00
    3. World'sFair1.00
    4. Multi-model1.00
    5. harnesses1.00
    6. Software copilots1.00
    7. Org-level harnesses1.00
    8. TRACK 6· JUNE 30, 20260.96
    9. World'sFair1.00
    10. Voice & Realtime Al0.99

  • 4:09 #12 done68 line(s)

    shot 12·sharpness 2022.4

    1. AlEngineer0.99
    2. 081.00
    3. World's Fair0.99
    4. Elle0.93
    5. WOO0.50
    6. Ele0.86
    7. Edit0.98
    8. Options0.98
    9. Navigate0.96
    10. Annotate Starting Points1.00
    11. NCSA Mosaic for MS Windows0.97
    12. Personal1.00
    13. Help0.98
    14. <body>0.97
    15. <base href=htspi//ww.yahoo.com/>0.83
    16. <html>0.87
    17. <head>0.99
    18. <title>Yahool</title>0.96
    19. </bead>0.95
    20. <form action=*http://search.yahoo.com/bin/search>0.95
    21. YAHOO!1.00
    22. <a href-htsp1//o.yahoo.com/homet/SnaceID-0:U/AdID-4833.0.85
    23. <a href/bin/tonl><IMc alt-[Yahoo1]" ismap vidth-460 he0.68
    24. <a hret-*/homet/7httpi//local.vaboo.com/local/">Get Local-0.85
    25. <1-- Parans 0, 4833 ->0.87
    26. <center>1.00
    27. <a href-"/homet/7httpi//w.yahoo.com/picka/*>Weekly Picki0.91
    28. 1.00.97
    29. Re SpX T A0.55
    30. Yeloe Pos Posie Reish - Ctr Man – Puk Outai et Scos0.51
    31. • Aats nd Humaites - Achectes hmturscly Lémoes0.51
    32. • Pusiness and Froomy(ta) Cmcs Ioamt Qand0.56
    33. • Compoters and Iatersela'l - Itrt Www Schret Meda0.56
    34. ● Edacatien sercts Ka12 Cade Eatrace0.52
    35. • Entertarsewt Xra Casl Lika Mones Maic Heor0.53
    36. ● Gevermeat 9 Eections Poloc (Xtra' Amoes Law Mcarr0.65
    37. • Heat ral Mehone Erss Decares Emes0.54
    38. • Refereace - Labrins Dictomarsa Phone Humbers0.61
    39. • News and Media[Xhaf - rent Evestt Magstoes TV Niewisiets0.62
    40. ut510.000.68
    41. WedidyPia0.56
    42. <a href=/search,html> Options</a>0.82
    43. <a href-hoen/lhai//munie.yaho.com>Stock Quotea</a>0.58
    44. <font size=-1>0.94
    45. <a href-"/homen/7httpu//yR.yaboo.com/yt.hm?FAM:yahookSEC=0.78
    46. <a href-*/homen/7httpi//ww.yahoo.com/search/people/>Peoj0.82
    47. <a hretomen/http//.vaboo.com/yaboo/>city ap0.58
    48. <a href-*/homen/7httpi//sports.yahoo.com>Sports Seores</0.88
    49. <a href-"Ata/Architecture/>Architecture</a>,0.79
    50. <a hret-"/Arta/Photosraphy/">Photography</a>,0.77
    51. <input size-30 name-p> <input type-submit value-Search>0.94
    52. <ul>0.96
    53. <11>0.98
    54. <a href-/Arta/><b>arts and Bumanities</b></a> -0.86
    55. </font>0.99
    56. </center>1.00
    57. <a href-"/Arta/Humanities/Literature/">Literature</a>...0.95
    58. ● Science - CS Paags Auong agseerng0.59
    59. • Serial Sdece Aatcosiss asake Eutis0.50
    60. • Society and Coltere Prole Eavcomeng Rrion0.65
    61. <a href"/Business_and Economy/Markets_and Inveatnenta>0.76
    62. <a hret=/headlines/buainesa/> <b>[xtrat |</b></a>-0.72
    63. <a href-"/Business_and_Economy/Conpanies/>Companies</a>,0.92
    64. <1>0.67
    65. <a href-"/Business_and_Economy/Classifieda/>Classifieds<0.93
    66. TRACK 6· JUNE 30, 20260.97
    67. World's Fair0.99
    68. Voice & Realtime Al0.99

  • 4:42 #13 done39 line(s)

    shot 13·sharpness 1916.8

    1. AlEngineer0.99
    2. 1995-20251.00
    3. 091.00
    4. World's Fair0.99
    5. Choose a ride0.99
    6. El0 050.69
    7. Elle Edi Oplions Mavigate Annotate Starting Points Personal0.93
    8. [0]0.62
    9. Holp0.86
    10. UberX 40.95
    11. 11:07 - 2 min away0.95
    12. Affordable rides,all to yourself0.90
    13. $10.851.00
    14. 11:07 -3 min away0.92
    15. Uber Green 41.00
    16. Eco-friendly1.00
    17. $10.851.00
    18. Comfort 40.99
    19. 11:07-3 min away0.93
    20. $14.241.00
    21. Newer cars with extra legroom0.99
    22. • Eetertatmeettad - Cad Lnia Mos Hanc Het0.52
    23. UberXL 60.98
    24. 11:08-4 min away0.97
    25. Affordablerides for groups up to 60.90
    26. $14.511.00
    27. • Reere d Spert( t es Tn As Qo0.56
    28. • Refereare - Lácons Dicsnes Pose Huchers0.61
    29. Comfort Electric 440.97
    30. 11:07-4 min away0.96
    31. $15.141.00
    32. Premium zero-emission cars1.00
    33. 1108-4 min away0.99
    34. Uber Pet 40.96
    35. $13.951.00
    36. Affordable rides for you and your pet1.00
    37. TRACK 6· JUNE 30, 20260.96
    38. Voice & Realtime Al0.98
    39. World's Fair1.00

  • 5:22 #14 done27 line(s)

    shot 14·sharpness 2171.3

    1. AlEngineer0.96
    2. World'sFair1.00
    3. 101.00
    4. Abacus1.00
    5. Keyboard accounting1.00
    6. machine1.00
    7. Arithmetical machine1.00
    8. Memex1.00
    9. o0.72
    10. 19451.00
    11. PRESENTED BY1.00
    12. well attempt to grasp the game of poker entirely by1.00
    13. the use of the mathematics of probability. The abacus,0.98
    14. Microsoft1.00
    15. with its beads strung on parallel wires, led the Arabs0.99
    16. to positional numeration and the concept of zero many0.99
    17. centuries before the rest of the world; and it was a0.99
    18. useful tool —so useful that it still exists.0.98
    19. It is a far cry from the abacus to the modern key-1.00
    20. board accounting machine. It will be an equal step1.00
    21. to the arithmetical machine of the future. But even1.00
    22. this new machine will not take the scientist where he0.99
    23. needs to go. Relief must be secured from laborious1.00
    24. detailed manipulation of higher mathematics as well,0.99
    25. TRACK 6· JUNE 30, 20260.97
    26. World'sFair1.00
    27. Voice & Realtime Al0.98

  • 5:51 #15 skipped

    shot 15·duplicate of #14

  • 6:08 #16 skipped

    shot 16·duplicate of #4

  • 6:30 #17 skipped

    shot 17·duplicate of #6

  • 7:03 #18 skipped

    shot 18·duplicate of #11

  • 7:25 #19 done9 line(s)

    shot 19·sharpness 1334.2

    1. AlEngineer0.99
    2. World'sFair1.00
    3. 19671.00
    4. 151.00
    5. Star Trek1.00
    6. GノQ0.64
    7. TRACK 6· JUNE 30, 20260.97
    8. World's Fair0.97
    9. Voice & Realtime Al0.97

  • 7:31 #20 done9 line(s)

    shot 20·sharpness 1344.4

    1. AlEngineer0.99
    2. World'sFair1.00
    3. 19671.00
    4. 151.00
    5. Star Trek1.00
    6. GQ0.63
    7. TRACK 6· JUNE 30, 20260.96
    8. World's Fair0.96
    9. Voice & Realtime Al0.98

  • 7:41 #21 done7 line(s)

    shot 21·sharpness 1356.3

    1. AlEngineer0.99
    2. World'sFair1.00
    3. 19671.00
    4. Star Trek1.00
    5. TRACK 6· JUNE 30, 20260.96
    6. World's Fair0.96
    7. Voice & Realtime Al1.00

  • 7:48 #22 done7 line(s)

    shot 22·sharpness 1281.0

    1. AlEngineer0.99
    2. 19671.00
    3. World'sFair0.97
    4. Star Trek1.00
    5. TRACK 6· JUNE 30, 20260.97
    6. World'sFair1.00
    7. Voice & Realtime Al0.98

  • 7:54 #23 done7 line(s)

    shot 23·sharpness 1305.1

    1. AlEngineer0.99
    2. 19671.00
    3. World'sFair1.00
    4. Star Trek1.00
    5. TRACK 6· JUNE 30, 20260.96
    6. World's Fair0.97
    7. Voice & Realtime Al0.98

Transcript

205 cues· 3,132 words· 17,706 chars

  1. 0:12 Good morning.
  2. 0:14 I know a lot of you in this room.
  3. 0:16 It's great to see you.
  4. 0:17 Welcome to the Voice Track at AI Engineer World's Fair.
  5. 0:20 For those of you who don't know me, my name is Quinla Holtmann-Kramer.
  6. 0:24 I work at a company called Daily.
  7. 0:26 We make developer infrastructure for real-time audio, video, and AI.
  8. 0:31 And we're the team behind Pipecat, which is the most widely used framework for building voice agents today.
  9. 0:37 Pipecat is open source and vendor neutral.
  10. 0:39 It's used by companies like AWS and Nvidia and Anthropic and thousands of startups and scale-ups and enterprises.
  11. 0:48 And today I'm gonna talk about what kind of agents we're building today, including voice agents, but not just voice agents, and what I'm interested in building next.
  12. 0:58 And I'm gonna try to put all this in the context of the roughly 80-year history of digital computing so far.
  13. 1:06 So we've got a lot to cover, we're gonna go fast.
  14. 1:09 But we're gonna start in 1945 with an essay called As We May Think written by an engineer, an academic, a civil servant named Vannevar Bush.
  15. 1:19 Bush deeply understood technologies ranging from analog computers to photography to radio to radar.
  16. 1:27 And as we may think is an extraordinary piece of writing, the essay predicts the development of, among other things, document display on a screen, and document scanning in OCR, and speech-to-text, and text-to-speech, and programming languages, and hypertext, and search engines, and data networks, something like the GoPro camera, something weirdly like the Amazon Kindle store, and voice interfaces, and brain computer interfaces.
  17. 1:53 And I've been thinking a lot about, as we may think lately, because Bush wrote this essay right at the very beginning of the computing age, and I think it feels to most of us like we're working right at the beginning of a new age, the intelligence age.
  18. 2:09 So what will we build?
  19. 2:13 Well, at the moment, we're building agents, and we're having a lot of fun doing it.
  20. 2:19 And a lot of the AI engineering work we're all talking about this week is focused on building a full, coherent software stack for AI agents.
  21. 2:28 Here is Satya Nadella talking a couple of weeks ago on a crossover episode of the no priors and latent space pod about the challenges of building agents in 2026.
  22. 2:37 That's right.
  23. 2:39 So in some sense, you kind of want the harness to define the models, the data, and the tools.
  24. 2:49 And so that you have a loop across those three.
  25. 2:51 And so what we are trying to, first of all, make sure is each of our products that we build, whether it's GitHub Copilot or the security stuff we showed with M-Dash or even the discovery for science, it doesn't matter.
  26. 3:04 All of them are multi-model harnesses with tools access so that you can do this progressive disclosure of tools even so that they're token efficient.
  27. 3:15 And then you're feeding it with very rich context.
  28. 3:20 So if you were here last year at AI Engineer World's Fair, you could draw a through line from the things we were talking about last year to loops and tool calls and context engineering and the stuff we're focused on this year to some emerging ideas.
  29. 3:34 You can hear that in Nadella's clip just there.
  30. 3:38 I think of this as kind of agents plus plus, like multi-agents.
  31. 3:41 model harnesses and software co-pilots embedded in every single piece of software, organization level harnesses.
  32. 3:47 So how do we go from Agents to Agents++ to the next thing beyond Agents?
  33. 3:54 Well, the last time we had this kind of massive change in how we write software and what we write software for and to do was the early days of the World Wide Web.
  34. 4:03 And I was around for the early days of the World Wide Web.
  35. 4:06 I was a baby programmer in 1995.
  36. 4:08 And the thing we talked about all the time in 1995, the way we talk about agents today, is web pages.
  37. 4:15 I spent a lot of time writing HTML by hand and building web server software in C and indexing and search software in C and authoring tooling and management infrastructure for web pages in Perl.
  38. 4:27 I was as excited about HTML in 1995 as I am about agents today.
  39. 4:34 And the web page is still with us and it's still important and useful, but today we talk a lot more about web applications and native mobile applications than we talk about web pages.
  40. 4:44 So just like we went from web pages to full blown web and native mobile, clearly we're gonna chart a path to a new fully AI native software that comes after Agents and Agents++.
  41. 4:57 So let's keep going back in time in order to think about this future.
  42. 5:02 Here's a timeline Vannevar Bush lays out in As We May Think.
  43. 5:05 He talks about the abacus, which was both an immensely useful device for doing practical, everyday mathematical calculations, and also an incredibly important theoretical tool that led to ideas like numeric place value and the concept of zero.
  44. 5:22 And Bush talks about the massive jump from the abacus to the state-of-the-art electromechanical keyboard calculating machines that he had in 1945.
  45. 5:33 And then he posits that we're about, or he is about to
  46. 5:40 witness and help create an equally large leap to what he calls the arithmetical machine.
  47. 5:48 And then he goes a step even further than that and he invents or designs in that essay a device he calls the mimics.
  48. 5:56 And we have a little bit of an advantage over Bush in 1945.
  49. 5:59 We've seen 80 years of computing play out.
  50. 6:02 So we can modify his timeline a little bit.

Chapters

  1. 0:00 Daily, Pipecat, and what comes after agents
  2. 1:31 What Vannevar Bush predicted in 1945
  3. 2:13 Nadella on multimodel harnesses
  4. 4:12 Agents are the web pages of 1995
  5. 5:25 From the abacus to the stored program computer
  6. 6:46 The 1950s: getting human intent into the machine
  7. 7:30 The 1960s: interactivity, Sketchpad, and Star Trek
  8. 8:51 The 1970s: abstractions that scale
  9. 9:32 VisiCalc and why it did not replace accountants
  10. 10:49 Knowledge Navigator, Apple's 1987 prediction
  11. 12:46 The web was multimodal from the start
  12. 13:24 Minority Report, Iron Man, and building it for real
  13. 17:10 Cloud, then agents, then what
  14. 18:24 Knowledge Navigator rebuilt on real technology
  15. 19:38 Gradient Bang and the primitives it exercises
  16. 20:52 Closing

Open at this second