vidtheque.

read-only demo

Videos vSx5IULvBns

Always-on agents run production without the on-call tax — Justin Smith, Resolve AI

index_state ready data_status ok

AI Engineer· published 2026-08-09· 0:24:56· en-US· indexed 2026-08-10 19:46

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 0:22, 1 of 1 keyframes kept
  5. Shot 4, 0:22 to 0:57, 1 of 1 keyframes kept
  6. Shot 5, 0:57 to 1:43, 1 of 1 keyframes kept
  7. Shot 6, 1:43 to 2:32, 1 of 1 keyframes kept
  8. Shot 7, 2:32 to 3:11, 1 of 1 keyframes kept
  9. Shot 8, 3:11 to 3:39, 1 of 1 keyframes kept
  10. Shot 9, 3:39 to 4:06, 1 of 1 keyframes kept
  11. Shot 10, 4:06 to 4:44, 1 of 1 keyframes kept
  12. Shot 11, 4:44 to 5:12, 1 of 1 keyframes kept
  13. Shot 12, 5:12 to 5:41, 0 of 1 keyframes kept
  14. Shot 13, 5:41 to 6:10, 0 of 1 keyframes kept
  15. Shot 14, 6:10 to 6:38, 0 of 1 keyframes kept
  16. Shot 15, 6:38 to 7:07, 0 of 1 keyframes kept
  17. Shot 16, 7:07 to 7:36, 0 of 1 keyframes kept
  18. Shot 17, 7:36 to 7:43, 1 of 1 keyframes kept
  19. Shot 18, 7:43 to 8:04, 1 of 1 keyframes kept
  20. Shot 19, 8:04 to 8:51, 1 of 1 keyframes kept
  21. Shot 20, 8:51 to 9:19, 1 of 1 keyframes kept
  22. Shot 21, 9:19 to 9:47, 0 of 1 keyframes kept
  23. Shot 22, 9:47 to 10:15, 0 of 1 keyframes kept
  24. Shot 23, 10:15 to 10:44, 1 of 1 keyframes kept
  25. Shot 24, 10:44 to 11:14, 0 of 1 keyframes kept
  26. Shot 25, 11:14 to 11:24, 1 of 1 keyframes kept
  27. Shot 26, 11:24 to 11:51, 1 of 1 keyframes kept
  28. Shot 27, 11:51 to 12:18, 0 of 1 keyframes kept
  29. Shot 28, 12:18 to 12:57, 1 of 1 keyframes kept
  30. Shot 29, 12:57 to 13:10, 0 of 1 keyframes kept
  31. Shot 30, 13:10 to 13:14, 1 of 1 keyframes kept
  32. Shot 31, 13:14 to 13:40, 0 of 1 keyframes kept
  33. Shot 32, 13:40 to 14:07, 0 of 1 keyframes kept
  34. Shot 33, 14:07 to 14:33, 1 of 1 keyframes kept
  35. Shot 34, 14:33 to 14:59, 0 of 1 keyframes kept
  36. Shot 35, 14:59 to 15:25, 0 of 1 keyframes kept
  37. Shot 36, 15:25 to 15:51, 0 of 1 keyframes kept
  38. Shot 37, 15:51 to 16:17, 0 of 1 keyframes kept
  39. Shot 38, 16:17 to 16:43, 0 of 1 keyframes kept
  40. Shot 39, 16:43 to 17:09, 0 of 1 keyframes kept
  41. Shot 40, 17:09 to 17:15, 1 of 1 keyframes kept
  42. Shot 41, 17:15 to 17:17, 1 of 1 keyframes kept
  43. Shot 42, 17:17 to 17:26, 1 of 1 keyframes kept
  44. Shot 43, 17:26 to 17:34, 1 of 1 keyframes kept
  45. Shot 44, 17:34 to 18:02, 1 of 1 keyframes kept
  46. Shot 45, 18:02 to 18:29, 0 of 1 keyframes kept
  47. Shot 46, 18:29 to 18:57, 0 of 1 keyframes kept
  48. Shot 47, 18:57 to 19:25, 0 of 1 keyframes kept
  49. Shot 48, 19:25 to 19:53, 0 of 1 keyframes kept
  50. Shot 49, 19:53 to 19:56, 1 of 1 keyframes kept
  51. Shot 50, 19:56 to 19:58, 0 of 1 keyframes kept
  52. Shot 51, 19:58 to 20:12, 0 of 1 keyframes kept
  53. Shot 52, 20:12 to 20:16, 0 of 1 keyframes kept
  54. Shot 53, 20:16 to 20:26, 1 of 1 keyframes kept
  55. Shot 54, 20:26 to 20:28, 1 of 1 keyframes kept
  56. Shot 55, 20:28 to 20:30, 1 of 1 keyframes kept
  57. Shot 56, 20:30 to 20:52, 1 of 1 keyframes kept
  58. Shot 57, 20:52 to 21:03, 1 of 1 keyframes kept
  59. Shot 58, 21:03 to 21:34, 1 of 1 keyframes kept
  60. Shot 59, 21:34 to 21:37, 1 of 1 keyframes kept
  61. Shot 60, 21:37 to 21:39, 1 of 1 keyframes kept
  62. Shot 61, 21:39 to 21:42, 1 of 1 keyframes kept
  63. Shot 62, 21:42 to 21:50, 1 of 1 keyframes kept
  64. Shot 63, 21:50 to 21:57, 0 of 1 keyframes kept
  65. Shot 64, 21:57 to 22:09, 0 of 1 keyframes kept
  66. Shot 65, 22:09 to 22:53, 0 of 1 keyframes kept
  67. Shot 66, 22:53 to 22:55, 1 of 1 keyframes kept
  68. Shot 67, 22:55 to 22:59, 1 of 1 keyframes kept
  69. Shot 68, 22:59 to 23:27, 1 of 1 keyframes kept
  70. Shot 69, 23:27 to 23:56, 0 of 1 keyframes kept
  71. Shot 70, 23:56 to 24:16, 1 of 1 keyframes kept
  72. Shot 71, 24:16 to 24:39, 1 of 1 keyframes kept
  73. Shot 72, 24:39 to 24:55, 0 of 1 keyframes kept

73 shot(s).

keyframes kept every frame deduplicated

What was stored

cues
307
whisperx 307
chunks
44
from 307 cues
keyframes
43
kept of 73 captured
frames with text
43
1,724 lines read
chapters
0
from the source metadata
keyframe bytes
11.0 MB
word timings on 307 cues

Provenance

Each pipeline stage, its state and the model that produced it
stage state model started took
fetch done — 2026-08-10 03:37 1m 42s
stt done — 2026-08-10 03:39 28s
chunk done — 2026-08-10 03:40 0s
text_embed done — 2026-08-10 19:46 1s
keyframe done — 2026-08-10 03:40 2m 43s
ocr done — 2026-08-10 03:42 30s
frame_embed done — 2026-08-10 19:46 7s

Frames, and what the machine read

  • 0:02 #0 done2 line(s)

    shot 0·sharpness 455.2

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

  • 0:03 #1 done2 line(s)

    shot 1·sharpness 665.1

    1. AlEngineer0.96
    2. World's Fair0.99

  • 0:10 #2 done24 line(s)

    shot 2·sharpness 2727.2

    1. LAB & PLATINUM SPONSORS0.99
    2. Amazon AGI Lab0.98
    3. ANTHROP\C1.00
    4. Google DeepMind1.00
    5. MINIMAX0.96
    6. OpenAI0.92
    7. Akamai1.00
    8. arize1.00
    9. aws1.00
    10. Braintrust bright data0.99
    11. B1.00
    12. Browserbase1.00
    13. docker1.00
    14. :neo4j0.91
    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

  • 0:21 #3 done3 line(s)

    shot 3·sharpness 237.8

    1. Resolve.ai1.00
    2. AlEngineer0.97
    3. World's Fair1.00

  • 0:53 #4 done14 line(s)

    shot 4·sharpness 2484.1

    1. AlEngineer0.97
    2. World'sFair1.00
    3. Justin Smith1.00
    4. Founding Product Engineer, Resolve Al1.00
    5. PRESENTED BY1.00
    6. Microsoft1.00
    7. 15+ years building developer tools and observability platforms0.99
    8. at companies like Splunk and VMware. Focused on product0.99
    9. design, frontend architecture, and turning complex1.00
    10. infrastructure data into usable experiences.0.99
    11. Resolve.ai1.00
    12. WorldsFalk0.88
    13. TRACK 8· JULY 2, 20260.93
    14. Agentic Engineering1.00

  • 1:11 #5 done16 line(s)

    shot 5·sharpness 2299.1

    1. AlEngineer0.98
    2. World'sFair1.00
    3. The first wave of AI changed0.99
    4. Bigger PRs0.99
    5. how we1.00
    6. build software1.00
    7. PRESENTED BY1.00
    8. More frequently1.00
    9. Microsoft1.00
    10. From developers (and non-developers?)0.99
    11. that might not know the code1.00
    12. BUT Developer productivity!1.00
    13. Resolve.ai1.00
    14. World's Fair0.85
    15. TRACK 8• JULY 2, 20260.95
    16. Agentic Engineering1.00

  • 2:26 #6 done32 line(s)

    shot 6·sharpness 2055.6

    1. AIE WF26 — Always-on age..0.83
    2. 2:25 PM - 2:45 PM0.94
    3. 60.70
    4. Take notes1.00
    5. AlEngineer0.96
    6. World'sFair1.00
    7. 70% of engineering time1.00
    8. Maintain platforms0.99
    9. Scale infra0.97
    10. is spent on1.00
    11. Debug1.00
    12. incidents1.00
    13. running software1.00
    14. aws1.00
    15. Ship hotfixes0.95
    16. Production1.00
    17. Alert1.00
    18. Coding1.00
    19. D0.84
    20. II0.66
    21. updates1.00
    22. Runbook1.00
    23. 00.52
    24. Patch vulnerabilities0.96
    25. Restore service0.98
    26. Escalations0.92
    27. Software Engineering1.00
    28. Resolve.ai1.00
    29. Spend Their Time? - doc #US53204750.98
    30. World's Falr0.84
    31. TRACK 8• JULY 2, 20260.96
    32. Agentic Engineering1.00

  • 2:59 #7 done54 line(s)

    shot 7·sharpness 4437.2

    1. Running software is getting1.00
    2. AIE WF26 — Always-on age..0.86
    3. 2:25 PM - 2:45 PM0.97
    4. Take notes0.99
    5. AlEngineer0.98
    6. World's Fair0.96
    7. exponentially harder1.00
    8. With Al generating majority of the code1.00
    9. AI IS CREATING LOT MORE ISSUES IN PRODUCTION0.98
    10. ERA OF "UNLIMITED TOKENS" IS COMING TO AN END0.98
    11. NEED FOR FULL STACK AI, NOT JUST MODELS1.00
    12. 1 in 5 organizations suffered a serious0.99
    13. incident linked to Al-generated code1.00
    14. Companies are scrambling1.00
    15. to curtail soaring AI costs0.99
    16. Prioritise Al Ecosystems, Not Just Al0.98
    17. Models: Microsoft CEO Satya Nadella0.99
    18. Yes amnor sue0.85
    19. 40%0.80
    20. tha0.52
    21. losing is appeal0.97
    22. building broader Al ecosystems, while emphasising that human capital will become even0.98
    23. The Microsoft CEO argued that the priority should shift from building frontier Al models to0.98
    24. more valuable in the Al era.0.99
    25. Jkn 85, 2026 15:21 pm 15T ①0.77
    26. Microsoft reports are exposing Al's1.00
    27. Why Domain-Specific AI Is Reshaping0.99
    28. more expensive than paying1.00
    29. real cost problem: Using the tech is1.00
    30. Enterprise Strategy1.00
    31. human employees0.97
    32. □ <0.56
    33. AI Coding Assistants Are Getting Worse0.99
    34. Newer models are more prone to silent but0.99
    35. Anthropic to disable its most advanced0.99
    36. deadly failure modes1.00
    37. Tokenminimizing: Meta1.00
    38. AI models after US order limiting foreign1.00
    39. Al-generated code contains more bugs and1.00
    40. Moves to Curb Employee AI0.99
    41. access1.00
    42. errors than human output0.99
    43. Usage as AI Costs Reach0.97
    44. By Craig Hale published December 18, 20250.93
    45. Al-generated code produces 1.7x more issues than human code0.99
    46. Billions1.00
    47. By yot Mann0.88
    48. Company said US government believes safeguards can be0.99
    49. bypassed and product used to identify software0.99
    50. vulnerabilities1.00
    51. Resolve.ai1.00
    52. World's Falr0.90
    53. TRACK 8• JULY 2, 20260.96
    54. Agentic Engineering0.98

  • 3:33 #8 done26 line(s)

    shot 8·sharpness 2203.8

    1. AIE WF26 — Ahways-on age...0.91
    2. 2:25 PM - 245 PM0.93
    3. 60.59
    4. Take notes1.00
    5. AlEngineer0.98
    6. World's Fair0.99
    7. Scale infra1.00
    8. Escalations1.00
    9. Why?1.00
    10. Alert response0.97
    11. Patck0.97
    12. vulnerabilities0.99
    13. Operating software in1.00
    14. production requires1.00
    15. Runbook updates1.00
    16. multiple teams1.00
    17. and systems0.99
    18. Ship1.00
    19. hotfixes1.00
    20. Restore service1.00
    21. Maintain platforms0.99
    22. Debug incidents0.97
    23. Resolve.ai1.00
    24. World's Fair0.89
    25. TRACK 8• JULY 2, 20260.96
    26. Agentic Engineering1.00

  • 3:58 #9 done8 line(s)

    shot 9·sharpness 1602.6

    1. AlEngineer0.99
    2. World'sFair1.00
    3. How do you fix this imbalance?1.00
    4. Al for Prod1.00
    5. Resolve.ai1.00
    6. Worid's Fair0.95
    7. TRACK 8• JULY 2, 20260.95
    8. Agentic Engineering1.00

  • 4:11 #10 done33 line(s)

    shot 10·sharpness 3319.4

    1. AlEngineer0.97
    2. World class engineering teams1.00
    3. World'sFair1.00
    4. using AI for prod1.00
    5. zscaler1.00
    6. DOORDASH1.00
    7. coinbase1.00
    8. fewer engineers in war rooms1.00
    9. Increased productivity - 30%1.00
    10. Increased reliability - 87%1.00
    11. faster investigations0.98
    12. debugging sessions weekly0.99
    13. Increased operational1.00
    14. hardening - 1oos of production0.97
    15. salesforce1.00
    16. MongoDB.1.00
    17. snowflake1.00
    18. expedia group1.00
    19. CISCO0.99
    20. toast0.97
    21. Fireworks Al0.95
    22. BLUEGROUND1.00
    23. MSCI0.96
    24. GGUIDEWIRE1.00
    25. i; veza°0.90
    26. ūpwind1.00
    27. GAMETIME1.00
    28. Pinecone1.00
    29. iFIT0.98
    30. Resolve.ai1.00
    31. World's Fair0.93
    32. TRACK 8• JULY 2, 20260.96
    33. Agentic Engineering1.00

  • 5:03 #11 done26 line(s)

    shot 11·sharpness 2153.8

    1. AI for prod1.00
    2. AlEngineer0.99
    3. World's Fair0.96
    4. AI AGENTS1.00
    5. + Al agents that handle on-call, incidents,0.99
    6. and daily operational tasks0.98
    7. On-call agents0.99
    8. Incidents agents1.00
    9. Background agents1.00
    10. Your agents1.00
    11. AGENT ARCHITECTURE1.00
    12. + Trusted to get to root cause in the most0.98
    13. demanding production environments1.00
    14. Models1.00
    15. Context · Reasoning0.96
    16. Actions1.00
    17. Learning1.00
    18. Evals1.00
    19. ENTERPRISE GRADE PLATFORM0.97
    20. + Fits into your agent ecosystem, or build1.00
    21. your agents on Resolve Al0.99
    22. Integrations · Cost efficiency · Agent Security · Usage & audit · Deployment0.96
    23. Resolve.ai1.00
    24. World's Fair0.93
    25. TRACK 8· JULY 2, 20260.93
    26. Agentic Engineering1.00

  • 5:37 #12 skipped

    shot 12·duplicate of #11

  • 5:52 #13 skipped

    shot 13·duplicate of #11

  • 6:13 #14 skipped

    shot 14·duplicate of #11

  • 7:03 #15 skipped

    shot 15·duplicate of #11

  • 7:10 #16 skipped

    shot 16·duplicate of #11

  • 7:40 #17 done63 line(s)

    shot 17·sharpness 2386.9

    1. Agents to run and fix your software0.98
    2. AlEngineer0.98
    3. World's Fair0.97
    4. AI AGENTS1.00
    5. On-call agent1.00
    6. Incidents agent0.98
    7. Background agents1.00
    8. Your agents0.96
    9. Autonomous triage1.00
    10. Shared investigation thread0.99
    11. Schedule on event or trigger1.00
    12. MCP/API/Skill0.99
    13. Root cause backed by evidence1.00
    14. Multiplayer investigation1.00
    15. Multi-turn chat1.00
    16. Drives incident channel1.00
    17. AI AGENT ARCHITECTURE1.00
    18. Evals1.00
    19. Models1.00
    20. Context1.00
    21. Reasoning1.00
    22. Actions1.00
    23. Learning1.00
    24. Absorb new model and1.00
    25. harness releases1.00
    26. Post-trained models1.00
    27. Frontier model1.00
    28. Knowledge graph1.00
    29. Tool fluency0.99
    30. Causal reasoning1.00
    31. Multi-agent1.00
    32. Guardrails0.99
    33. Scoped autonomy0.98
    34. Implicit signals1.00
    35. Explicit feedback0.98
    36. Domain specialized eval0.99
    37. framework1.00
    38. orchestration1.00
    39. Context engineering1.00
    40. coordination1.00
    41. Calibrated to mirror1.00
    42. engineer's investigation0.99
    43. ENTERPRISE GRADE PLATFORM1.00
    44. Integrations1.00
    45. Cost efficiency1.00
    46. Agent Security1.00
    47. Usage & audit0.97
    48. Deployments1.00
    49. 60+ connectors0.99
    50. Optimized token usage1.00
    51. Tenant isolation1.00
    52. Usage reports0.98
    53. Cloud0.92
    54. Custom tool integrations0.99
    55. Credit based pricing0.96
    56. Data protection1.00
    57. Trace agent reasoning1.00
    58. Resolve AI Managed VPC0.99
    59. w/BYOK*0.97
    60. Resolve.ai1.00
    61. Worid's Falr0.87
    62. TRACK 8· JULY 2, 20260.95
    63. Agentic Engineering1.00

  • 7:49 #18 done60 line(s)

    shot 18·sharpness 1592.0

    1. Agents to run and fix your software0.99
    2. AlEngineer0.96
    3. World's Fair0.99
    4. AIAGENTS0.99
    5. On-call agent0.98
    6. Incidents agent1.00
    7. Background agents0.99
    8. Your agents1.00
    9. Autonomous triage1.00
    10. Shared investigation thread0.99
    11. O Schedule on event or trigger0.97
    12. MCP/API/Skill0.99
    13. Root cause backed by evidence1.00
    14. Multiplayer investigation0.98
    15. Multi-turn chat0.97
    16. Drives incident channel0.98
    17. AI AGENT ARCHITECTURE0.97
    18. Evals1.00
    19. Models1.00
    20. Context1.00
    21. Reasoning1.00
    22. Actions1.00
    23. Learning1.00
    24. harness releases1.00
    25. Absorb new model and0.99
    26. Frontier model1.00
    27. Tool fluency1.00
    28. Multi-agent0.96
    29. Scoped autonomy1.00
    30. Guardrails0.99
    31. Explicit feedback1.00
    32. Implicit signals1.00
    33. Domain specialized eval0.98
    34. framework1.00
    35. orchestration1.00
    36. Context engineering0.97
    37. coordination0.98
    38. engineer's investigation1.00
    39. Calibrated to mirror0.99
    40. ENTERPRISE GRADE PLATFORM0.98
    41. Integrations1.00
    42. Cost efficiency1.00
    43. Agent Security1.00
    44. Usage & audit1.00
    45. Deployments1.00
    46. 60+ connectors0.99
    47. Optimized token usage0.99
    48. Tenant isolation1.00
    49. Usage reports1.00
    50. Cloud1.00
    51. Custom tool integrations0.99
    52. Credit based pricing1.00
    53. Data protection0.98
    54. Trace agent reasoning1.00
    55. Resolve AI Managed VPC0.97
    56. w/BYOK *0.90
    57. Resolve.ai1.00
    58. World's Fair0.90
    59. TRACK 8• JULY 2, 20260.95
    60. Agentic Engineering1.00

  • 8:09 #19 done9 line(s)

    shot 19·sharpness 1700.3

    1. AlEngineer0.98
    2. World's Fair0.96
    3. LIVE POLL · SHOW OF HANDS0.96
    4. Are you using0.98
    5. agents in your1.00
    6. daily work today?1.00
    7. Resolve.ai1.00
    8. TRACK 8• JULY 2, 20260.96
    9. Agentic Engineering1.00

  • 9:02 #20 done19 line(s)

    shot 20·sharpness 2897.5

    1. AlEngineer0.99
    2. Most production work is always there and0.98
    3. World'sFair1.00
    4. not always α critical event0.98
    5. Watch the deploy that just went out0.99
    6. On-call has a page. Incidents have a bridge.1.00
    7. Write the morning deploy and incident digest1.00
    8. INVISIBLE TOIL0.98
    9. Re-investigate the p99 drift that came back1.00
    10. This work has no critical1.00
    11. trigger that makes anyone1.00
    12. Produce the weekly capacity report1.00
    13. account for it, so it stays1.00
    14. invisible and adds up.1.00
    15. Run the recurring health check nobody owns1.00
    16. Resolve.ai1.00
    17. Worid's Fair0.95
    18. TRACK 8· JULY 2, 20260.95
    19. Agentic Engineering1.00

  • 9:25 #21 skipped

    shot 21·duplicate of #20

  • 10:01 #22 skipped

    shot 22·duplicate of #20

  • 10:35 #23 done21 line(s)

    shot 23·sharpness 3679.3

    1. Every operational task requires1.00
    2. AlEngineer0.99
    3. World'sFair1.00
    4. two things0.96
    5. Task = Execution × Production Context0.98
    6. PRESENTED BY1.00
    7. Microsoft1.00
    8. Execution1.00
    9. Production Context1.00
    10. The task in isolation, the actual analysis or change.1.00
    11. What you know about the specific environment the1.00
    12. Small and roughly fixed.0.99
    13. work is being done in to know how to execute and1.00
    14. evaluate the work being done. Knowledge and tools.1.00
    15. Large and growing.0.99
    16. Production Context » Execution.0.99
    17. Navigating your environment is a major contributor to toil.0.99
    18. Resolve.ai1.00
    19. Worid's Fair0.89
    20. TRACK 8• JULY 2, 20260.95
    21. Agentic Engineering1.00

Transcript

307 cues· 4,711 words· 24,653 chars

  1. 0:12 Hello, hello.
  2. 0:15 Hey everybody, welcome to this talk, always on agents run production without the on-call text.
  3. 0:23 My name is Justin Smith, one of the founding product engineers at Resolve AI.
  4. 0:28 Been in the space for about 15 plus years in the sort of monitoring, observability, how do you kind of operate production systems space.
  5. 0:37 Was at Splunk for a while, was one of the architects on the observability suite there.
  6. 0:42 I spent a good tenure at VMware.
  7. 0:45 Really, really enjoy product design and front end architecture.
  8. 0:48 How do people experience a product or a use case or something like that?
  9. 0:54 That's the stuff I like to dabble in.
  10. 1:00 But I want to talk a little bit about the first wave of AI.
  11. 1:04 And it's been a fun one.
  12. 1:06 I think the first big wave, and I'm sure we've all experienced this, is just how we build software.
  13. 1:13 But there's some sort of net effects of that.
  14. 1:15 It's a lot of bigger PRs that are coming through.
  15. 1:18 We definitely see a lot of this a lot more frequently.
  16. 1:21 So people are shipping code at a much faster rate from developers and we're beginning to see maybe from even non-developers that maybe don't actually know the code or what it's doing or the sort of like operating principles behind it.
  17. 1:35 But we're getting developer productivity
  18. 1:39 And that's good, right?
  19. 1:40 That's a good thing that we're all able to produce more and faster.
  20. 1:45 kind of, sort of, what we actually found out, and this was a survey study done, is that 70% of the time from an engineer is actually not just like, is not focused just on writing code.
  21. 1:56 It's actually spent on actually running the code that is actually shipped into production.
  22. 2:00 Maintaining all the platforms, scaling the infrastructure, debugging all the incidents and being on call, shipping hot fixes, right, dealing with alerts,
  23. 2:10 updating all the sort of run books and operating procedures, restoring services, dealing with escalations, dealing with sort of like questions from other teams and things like that.
  24. 2:20 So really coding was never the big bottleneck, right?
  25. 2:25 A lot of it was really around, thank you, Granola.
  26. 2:28 A lot of it was really around like how do we actually run these things sort of in production?
  27. 2:34 And that's getting harder and harder and harder.
  28. 2:37 AI is creating a lot more issues in production as AI code sort of goes through.
  29. 2:43 It's not clear we have the right sort of structures in place to deal with the amount of kind of changes that are coming through.
  30. 2:49 Unlimited tokens is sort of coming to an end.
  31. 2:51 The token max, right, they're starting to clamp down.
  32. 2:54 Prices are going up.
  33. 2:55 Companies are getting a lot more stringent on what's being used for AI.
  34. 3:00 We need full stack AI.
  35. 3:01 It's not just about the models anymore.
  36. 3:03 It's about the context around the models and what the models can do inside of a specific domain.
  37. 3:08 These become the problem areas that we need to sort of focus and tackle on.
  38. 3:13 And this is true today, right?
  39. 3:15 So it's creating more sort of complexity inside of our environment.
  40. 3:19 I mean, the reality is that systems have always been complex.
  41. 3:22 That's why we have these big tools that can try to give us insights into these systems.
  42. 3:28 There are multiple teams, there's multiple systems that are all having to work together and they all have their own goals that they're trying to deliver towards, but you have organizational goals.
  43. 3:37 And how do you keep all of this sort of in
  44. 3:42 Let me get rid of this.
  45. 3:43 How do you keep all of this in balance, right?
  46. 3:46 How do you pull all of this stuff together in a way that actually helps you and facilitates your organization?
  47. 3:55 And the answer is, well, you gotta use AI inside of production to deal with sort of the amount of increase of complexity that AI is kind of putting into your product or into your system.
  48. 4:08 And so that's where Resolve, you know, this was kind of our sort of hypothesis from the beginning was, you know, we're going to see an influx in, you know, issues coming out of coding, just the increase in coding velocity.
  49. 4:22 There's going to be more need for kind of AI to actually operate and run these systems.
  50. 4:26 We're, you know, lucky to work with, you know, some world-class engineering teams that are solving like really difficult problems at

Open at this second