read-only demo

Videos Btk8wDUVs74

From Systems of Record to Systems of Context — Omri Bruchim & Tomer Ast, monday.com

index_state ready data_status ok

AI Engineer· published 2026-07-22· 0:15:57· en-US· indexed 2026-08-11 00:27

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 0:46, 1 of 1 keyframes kept
  6. Shot 5, 0:46 to 1:11, 0 of 1 keyframes kept
  7. Shot 6, 1:11 to 1:37, 1 of 1 keyframes kept
  8. Shot 7, 1:37 to 2:02, 1 of 1 keyframes kept
  9. Shot 8, 2:02 to 2:28, 1 of 1 keyframes kept
  10. Shot 9, 2:28 to 2:53, 1 of 1 keyframes kept
  11. Shot 10, 2:53 to 3:19, 1 of 1 keyframes kept
  12. Shot 11, 3:19 to 3:45, 1 of 1 keyframes kept
  13. Shot 12, 3:45 to 4:10, 1 of 1 keyframes kept
  14. Shot 13, 4:10 to 4:36, 0 of 1 keyframes kept
  15. Shot 14, 4:36 to 5:01, 1 of 1 keyframes kept
  16. Shot 15, 5:01 to 5:27, 1 of 1 keyframes kept
  17. Shot 16, 5:27 to 5:52, 1 of 1 keyframes kept
  18. Shot 17, 5:52 to 6:18, 0 of 1 keyframes kept
  19. Shot 18, 6:18 to 6:44, 1 of 1 keyframes kept
  20. Shot 19, 6:44 to 7:09, 0 of 1 keyframes kept
  21. Shot 20, 7:09 to 7:44, 1 of 1 keyframes kept
  22. Shot 21, 7:44 to 8:13, 1 of 1 keyframes kept
  23. Shot 22, 8:13 to 8:38, 1 of 1 keyframes kept
  24. Shot 23, 8:38 to 9:04, 0 of 1 keyframes kept
  25. Shot 24, 9:04 to 9:23, 1 of 1 keyframes kept
  26. Shot 25, 9:23 to 9:54, 1 of 1 keyframes kept
  27. Shot 26, 9:54 to 10:14, 0 of 1 keyframes kept
  28. Shot 27, 10:14 to 10:34, 1 of 1 keyframes kept
  29. Shot 28, 10:34 to 10:49, 1 of 1 keyframes kept
  30. Shot 29, 10:49 to 11:12, 1 of 1 keyframes kept
  31. Shot 30, 11:12 to 11:52, 1 of 1 keyframes kept
  32. Shot 31, 11:52 to 12:29, 1 of 1 keyframes kept
  33. Shot 32, 12:29 to 12:54, 1 of 1 keyframes kept
  34. Shot 33, 12:54 to 13:23, 1 of 1 keyframes kept
  35. Shot 34, 13:23 to 13:55, 1 of 1 keyframes kept
  36. Shot 35, 13:55 to 14:27, 0 of 1 keyframes kept
  37. Shot 36, 14:27 to 15:00, 0 of 1 keyframes kept
  38. Shot 37, 15:00 to 15:22, 1 of 1 keyframes kept
  39. Shot 38, 15:22 to 15:32, 1 of 1 keyframes kept
  40. Shot 39, 15:32 to 15:40, 1 of 1 keyframes kept
  41. Shot 40, 15:40 to 15:57, 0 of 1 keyframes kept

41 shot(s).

keyframes kept every frame deduplicated

What was stored

cues
196
whisperx 196
chunks
28
from 196 cues
keyframes
32
kept of 41 captured
frames with text
32
691 lines read
chapters
9
from the source metadata
keyframe bytes
5.0 MB
word timings on 196 cues

Provenance

Each pipeline stage, its state and the model that produced it
stage state model started took
fetch done 2026-08-11 00:24 1m 07s
stt done 2026-08-11 00:25 16s
chunk done 2026-08-11 00:25 0s
text_embed done 2026-08-11 00:25 0s
keyframe done 2026-08-11 00:25 1m 42s
ocr done 2026-08-11 00:27 15s
frame_embed done 2026-08-11 00:27 6s

Frames, and what the machine read

  • 0:02 #0 done2 line(s)

    shot 0·sharpness 449.7

    1. AlEngineer0.96
    2. World's Fair1.00
  • 0:03 #1 done2 line(s)

    shot 1·sharpness 662.0

    1. AIEngineer0.95
    2. World's Fair0.99
  • 0:10 #2 done24 line(s)

    shot 2·sharpness 2739.9

    1. LAB & PLATINUM SPONSORS0.99
    2. Amazon AGI Lab0.98
    3. ANTHROP\C1.00
    4. Google DeepMind1.00
    5. MINIMAX0.97
    6. OpenAI0.92
    7. Akamai1.00
    8. arize0.92
    9. aws1.00
    10. Braintrust bright data0.98
    11. B1.00
    12. Browserbase1.00
    13. docker1.00
    14. :neo4j0.93
    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
  • 0:19 #3 done2 line(s)

    shot 3·sharpness 316.6

    1. AlEngineer1.00
    2. World'sFair1.00
  • 0:40 #4 done13 line(s)

    shot 4·sharpness 2023.2

    1. AlEngineer0.98
    2. World's Fair0.98
    3. .monday Al engineering0.98
    4. From Systems of0.99
    5. PRESENTED BY1.00
    6. Records to Systems1.00
    7. Microsoft1.00
    8. of Context1.00
    9. Tomer Ast1.00
    10. Omri Bruchim0.99
    11. 1/ 300.92
    12. Workd's Fair0.96
    13. Engineering the future of Al0.99
  • 0:49 #5 skipped

    shot 5·duplicate of #4

  • 1:14 #6 done14 line(s)

    shot 6·sharpness 1494.6

    1. Edi10.84
    2. AlEngineer0.98
    3. World's Fair0.97
    4. 661.00
    5. What Should I0.99
    6. Focus on Right1.00
    7. PRESENTED BY1.00
    8. Microsoft1.00
    9. Now?|0.88
    10. 40.92
    11. 2 /300.91
    12. Workd's Fair0.94
    13. TRACK 5· JULY 2, 20260.95
    14. Graphs1.00
  • 1:52 #7 done43 line(s)

    shot 7·sharpness 1673.5

    1. Edit0.98
    2. AlEngineer0.99
    3. Your Al assistant has every board,0.99
    4. World's Fair0.99
    5. task, email, and message youve0.99
    6. ever touched.0.98
    7. And it still can't really answer.0.97
    8. Inbox1.00
    9. //. Team Projects0.95
    10. Re: launch services agreement0.98
    11. SpaceX0.94
    12. 2.4 AM0.76
    13. VP sync · 9:00-9:300.94
    14. Calendar1.00
    15. Buy domain0.96
    16. Dana0.99
    17. coall0.90
    18. New website1.00
    19. Hi, when wil my order bee0.93
    20. shipped?1.00
    21. #4821 Launch0.95
    22. Your order has shippedl Aeriving0.96
    23. May 50.96
    24. Notes - Al Engineer conf0.95
    25. ##launch0.95
    26. Open on the stil cant0.89
    27. answer hook0.96
    28. Records meaning0.97
    29. Acme Corp · reports to0.96
    30. VP1.00
    31. Org chart0.99
    32. Tomer Ast0.99
    33. Hey, do you remember to send me0.97
    34. the slide for the meeting trw0.95
    35. morning?0.99
    36. 1UIS PM0.62
    37. Slow + fast engines0.91
    38. world model0.99
    39. .monday Al engineering0.97
    40. 3 / 300.94
    41. Workd's Fair0.89
    42. TRACK 5· JULY 2, 20260.95
    43. Graphs1.00
  • 2:25 #8 done11 line(s)

    shot 8·sharpness 1794.1

    1. Edit0.94
    2. AlEngineer0.98
    3. World's Fair0.97
    4. The Challenge1.00
    5. Not missing data.0.97
    6. Missing understanding!1.00
    7. 40.95
    8. 4 / 300.86
    9. Workd's Fair0.90
    10. TRACK 5• JULY 2, 20260.96
    11. Graphs1.00
  • 2:45 #9 done26 line(s)

    shot 9·sharpness 2160.5

    1. Edit0.99
    2. AlEngineer0.97
    3. .monday.com0.97
    4. World'sFair1.00
    5. A0.53
    6. Marketing1.00
    7. Projects1.00
    8. Sidekick1.00
    9. monday.com is a global software company1.00
    10. that helps customers scale business impact1.00
    11. and expand what teams can deliver.1.00
    12. Vibe1.00
    13. Operations1.00
    14. Product1.00
    15. The best0.99
    16. Design1.00
    17. Agents1.00
    18. Al work0.95
    19. 880.86
    20. platform1.00
    21. CRM1.00
    22. HR1.00
    23. 6 / 300.84
    24. Worid's Fair0.93
    25. TRACK 5• JULY 2, 20260.95
    26. Graphs1.00
  • 3:13 #10 done34 line(s)

    shot 10·sharpness 2245.8

    1. Our mission has always been to help teams1.00
    2. Edit0.88
    3. AlEngineer0.98
    4. World's Fair0.99
    5. achieve their business outcomes0.99
    6. Sales1.00
    7. Marketing1.00
    8. Accelerate revenue1.00
    9. Accelerate demand1.00
    10. and growth1.00
    11. and pipeline growth1.00
    12. Operations1.00
    13. Finance1.00
    14. Maximize efficiency0.98
    15. Streamline financial1.00
    16. and workflow agility0.99
    17. operations efficiently0.99
    18. 40.94
    19. Support0.99
    20. Legal1.00
    21. Provide service at0.99
    22. Minimize risk and0.99
    23. speed and scale0.98
    24. accelerate cycles1.00
    25. Product & R&D0.99
    26. HR1.00
    27. Speed up innovation1.00
    28. Optimize the1.00
    29. and build anything1.00
    30. employee journev0.99
    31. 7 / 300.89
    32. Worid's Fair0.93
    33. TRACK 5· JULY 2, 20260.94
    34. Graphs1.00
  • 3:39 #11 done44 line(s)

    shot 11·sharpness 2548.4

    1. Al work capabilities0.99
    2. Notes1.00
    3. Edit0.89
    4. AlEngineer0.99
    5. World's Fair0.99
    6. monday1.00
    7. monday1.00
    8. monday1.00
    9. monday1.00
    10. sidekick1.00
    11. vibe1.00
    12. agents1.00
    13. workflows1.00
    14. Exponential1.00
    15. Infinite1.00
    16. Unlimited1.00
    17. Agentic1.00
    18. productivity1.00
    19. software1.00
    20. workforce1.00
    21. workflows1.00
    22. Your intelligent Al personal1.00
    23. Turn any business need into a0.99
    24. Unlimited resources of expert1.00
    25. Orchestrate your work and business1.00
    26. assistant that understands your1.00
    27. complete solution - consolidate0.99
    28. agents doing the work for you1.00
    29. processes with Al1.00
    30. work, thinks and executes with you1.00
    31. your stack with just a prompt1.00
    32. Research competitors1.00
    33. Genera1.00
    34. Brainstorm idel0.95
    35. monday sidekick1.00
    36. Turn your words into0.99
    37. work apps0.99
    38. Tell sidekick what you want to do..0.99
    39. @0.99
    40. 40.59
    41. 8 / 300.81
    42. World's Fair0.91
    43. TRACK 5• JULY 2, 20260.97
    44. Graphs1.00
  • 4:07 #12 done32 line(s)

    shot 12·sharpness 2664.9

    1. AlEngineer0.99
    2. monday sidekick0.99
    3. World'sFair1.00
    4. Your intelligent Al personal assistant1.00
    5. that understands your work, thinks0.99
    6. and executes with you1.00
    7. 80.55
    8. Knows you and your business1.00
    9. Hey Sandra, ready to get started?1.00
    10. 0o0.81
    11. Describe what you need or ask a question..0.98
    12. Accelerates work at every level1.00
    13. 60.59
    14. @Mention assets0.98
    15. Works the way you do0.99
    16. Write a doc0.95
    17. Create a board0.99
    18. Analyze data0.99
    19. Generate an image1.00
    20. Buld a Vibe app0.93
    21. Keeps you in total control1.00
    22. Explore ready-made prompts for you0.99
    23. tuild an RSVP help event0.97
    24. Delivers answers you can trust1.00
    25. Omal0.95
    26. V Vibe0.95
    27. structures1.00
    28. Agerts0.84
    29. 9 / 300.89
    30. World's Fair0.91
    31. TRACK 5• JULY 2, 20260.95
    32. Graphs1.00
  • 4:13 #13 skipped

    shot 13·duplicate of #12

  • 4:58 #14 done44 line(s)

    shot 14·sharpness 1088.2

    1. Notes0.93
    2. Eait0.85
    3. AlEngineer0.98
    4. World's Fair0.96
    5. Q3 Launch0.98
    6. PRD v20.97
    7. #Launch0.96
    8. VP sync0.88
    9. Standup 9:000.95
    10. ● Sales· Acme0.90
    11. Unblock #48211.00
    12. Auth spec0.99
    13. #eng-oncall1.00
    14. THE CHALLENGE WE'RE FACING1.00
    15. Escalation1.00
    16. 1:11.00
    17. Bug triage1.00
    18. One assistant1.00
    19. Design notes0.99
    20. Retro notes0.96
    21. all of your1.00
    22. Ship hotfix1.00
    23. Re: contract0.98
    24. work.1.00
    25. Board mtg0.97
    26. Launch brief1.00
    27. 40.98
    28. #general1.00
    29. Renewal1.00
    30. Kickoff call1.00
    31. Send notes1.00
    32. Hiring1.00
    33. Boards1.00
    34. Docunents0.99
    35. ●Slack0.93
    36. Email1.00
    37. Calendar0.99
    38. ●Transcripts0.97
    39. Notetaker actions1.00
    40. .monday Al engineering0.97
    41. 10 / 300.90
    42. World's Fair0.91
    43. TRACK 5· JULY 2, 20260.95
    44. Graphs1.00
  • 5:21 #15 done20 line(s)

    shot 15·sharpness 1765.8

    1. Edit0.82
    2. AlEngineer0.97
    3. World'sFair1.00
    4. THE AGENT GAP1.00
    5. Sharp at doing tasks.0.99
    6. PRESENTED BY0.99
    7. 010.99
    8. Lost at finding them.0.98
    9. Microsoft1.00
    10. Point an agent at a problem and it's brilliant. But "what0.99
    11. should I focus on?" isn't a problem to solve — it's asking0.99
    12. the agent to find what the problems even are.0.99
    13. 40.99
    14. "Draft a reply to this escalation" → nails it.0.99
    15. "What should I work on first?" → guesses.0.97
    16. monday Al engineering1.00
    17. 11 / 300.90
    18. Workd's Fair0.86
    19. TRACK 5• JULY 2, 20260.95
    20. Graphs1.00
  • 5:47 #16 done22 line(s)

    shot 16·sharpness 1883.5

    1. AlEngineer0.98
    2. World'sFair1.00
    3. RECORDS ≠ MEANING1.00
    4. A log says what happened.0.99
    5. Never what it means.0.99
    6. 021.00
    7. PRESENTED BY1.00
    8. A record is like a line of code someone wrote months0.97
    9. Microsoft1.00
    10. ago. It tells you what changed — never why. To recover0.97
    11. the meaning you trace it back, one hop at a time.0.99
    12. The log: launch. due += 5d - what changed.0.96
    13. 40.99
    14. The why lives a hop away, every time:1.00
    15. git blame → the commit message → the PR description0.99
    16. → the monday task that asked for it → the customer who0.99
    17. complained in the first place.1.00
    18. monday Al engineering0.98
    19. 12 / 300.89
    20. Worid'sFair0.92
    21. TRACK 5• JULY 2, 20260.96
    22. Graphs1.00
  • 6:15 #17 skipped

    shot 17·duplicate of #15

  • 6:28 #18 done21 line(s)

    shot 18·sharpness 1821.6

    1. Edit1.00
    2. AlEngineer0.99
    3. World'sFair1.00
    4. RECORDS ≠ MEANING1.00
    5. A log says what happened.0.99
    6. Never what it means.0.99
    7. 021.00
    8. A record is like a line of code someone wrote months1.00
    9. ago. It tells you what changed — never why. To recover0.97
    10. the meaning you trace it back, one hop at a time.0.99
    11. The log: launch. due += 5d - what changed.0.97
    12. 40.99
    13. The why lives a hop away, every time:1.00
    14. git blame → the commit message → the PR description0.99
    15. → the monday task that asked for it → the customer who0.99
    16. complained in the first place.1.00
    17. monday Al engineering0.99
    18. 12 / 300.92
    19. World's Fair0.90
    20. TRACK 5• JULY 2, 20260.95
    21. Graphs1.00
  • 7:06 #19 skipped

    shot 19·duplicate of #18

  • 7:33 #20 done20 line(s)

    shot 20·sharpness 1702.7

    1. Edit0.99
    2. AlEngineer0.98
    3. World's Fair0.96
    4. WHAT WE BUILT0.95
    5. Not a bigger prompt.0.99
    6. The monday.com world0.99
    7. model.1.00
    8. why this matters1.00
    9. how to help1.00
    10. when - & when not - to act0.98
    11. Context that follows your work: who you are, what1.00
    12. 40.95
    13. you're working on, what needs action – a model the0.97
    14. agent can reason over to answer what it can't derive1.00
    15. on its own.0.99
    16. monday Al engineering0.98
    17. 14 / 300.90
    18. Worid's Fair0.95
    19. TRACK 5· JULY 2, 20260.95
    20. Graphs1.00
  • 8:07 #21 done14 line(s)

    shot 21·sharpness 1149.3

    1. Edit1.00
    2. AlEngineer0.98
    3. World'sFair1.00
    4. FIRST, WHAT IT'S NOT1.00
    5. It isn't retrieval.1.00
    6. The problem was never getting the data - hand an0.98
    7. agent every board, every doc, every message, and it's0.99
    8. still lost. So the model isn't a smarter search over your0.98
    9. work. It's a working understanding of it.0.98
    10. .monday Al engineering0.97
    11. 15 / 300.92
    12. World'sFair0.90
    13. TRACK 5· JULY 2, 20260.93
    14. Graphs1.00
  • 8:30 #22 done30 line(s)

    shot 22·sharpness 2083.4

    1. AlEngineer0.98
    2. World's Fair0.97
    3. WHAT THE MODEL UNDERSTANDS1.00
    4. Three faculties of understanding.1.00
    5. x0.55
    6. HOW IT'S STRUCTURED0.99
    7. WHAT'S TRUE RIGHT NOW0.96
    8. WHAT IT'S LEARNED1.00
    9. Structure & connections1.00
    10. The present0.96
    11. Learned over time1.00
    12. Items, people, meetings, boards -0.98
    13. Live signals read off your current1.00
    14. Episodes, decisions and their1.00
    15. and the relationships that carry the1.00
    16. state: what's overdue, what's piling0.99
    17. outcomes - distilled into durable0.97
    18. meaning: what blocks what, who's1.00
    19. into a crunch, who you're suddenly1.00
    20. patterns: how you actually work, not1.00
    21. waiting on whom, where a thing got0.99
    22. in the loop with.1.00
    23. how a job description says you do.1.00
    24. decided.1.00
    25. every fact stamped in time - what's true, what was true, and when it changed0.99
    26. monday Al engineering0.98
    27. 16/ 300.89
    28. Worid's Fair0.93
    29. TRACK 5· JULY 2,20260.95
    30. Graphs1.00
  • 8:44 #23 skipped

    shot 23·duplicate of #22

Transcript

196 cues· 2,367 words· 12,829 chars

  1. 0:14 Hey, everyone.
  2. 0:16 We are super excited to be here.
  3. 0:17 Thanks for having us.
  4. 0:19 Today, we're going to talk about how we shift Monday.com from a system of record into a system of context.
  5. 0:27 And honestly, the title tells the whole story in just a single line.
  6. 0:32 For decades, we built software that records what happened,
  7. 0:37 Every task, every document, every message, every status update just put into the record.
  8. 0:45 What we want to talk today is take it a step further.
  9. 0:48 We want software that actually understands the connection between them.
  10. 0:55 I want to start from a simple question that each one of us ask himself every morning.
  11. 1:00 What should I focus on right now?
  12. 1:03 It sounds almost trivial, but to be honest with yourself, if you ask your agent, whether it's Gemini, GPT, or even Cloud, if you ever typed this question, you probably got a list of bullets not related to each other, a list of items dressed up like a confident paragraph.
  13. 1:25 But it's not really connected to what you're working on.
  14. 1:30 Actually, I tested last week, and Claude asked me to go to the gym.
  15. 1:34 I don't know if it's a compliment or not, but this is what he suggested.
  16. 1:41 And what really make it so frustrating is like your AI assistant have all this data.
  17. 1:47 It has all the boards, the tasks, the emails, the Slack messages, everything that you ever touched if you connect it.
  18. 1:54 But it still can't really answer it.
  19. 1:58 It has all the data, but it has zero understanding.
  20. 2:04 So this is the real challenge we are facing.
  21. 2:07 This is the heart of the entire talk.
  22. 2:09 The problem was never the missing of data, the retrieval.
  23. 2:12 The problem is like the missing understanding.
  24. 2:15 Those are two totally different things and almost everyone mixed between them.
  25. 2:21 Understanding is the word that we're gonna focus the entire talk.
  26. 2:25 Not context, not memory, not retrieval, understanding.
  27. 2:30 So quick about our results.
  28. 2:32 My name is Tomer.
  29. 2:33 My name is Omri.
  30. 2:34 This is Tomer.
  31. 2:35 We're both engineering manager at Monday.com, working on exactly the problem that we're going to talk about.
  32. 2:43 A little bit context about where we're coming from.
  33. 2:46 Monday.com is a global software company.
  34. 2:49 We build a AI work platform used by hundreds of thousands of teams.
  35. 2:54 And the part that really matters for the talk is that Monday is where the work lives.
  36. 3:00 We help companies to doing the work, not just like saving records.
  37. 3:04 Every project, every task, every decision, every meeting, the nodes, the action item, everything logs into the system.
  38. 3:12 And our mission has always been to help the teams to achieve their business outcome.
  39. 3:18 Whether you are a salesperson, so we want to help you create an agent that's called to your prospect and help you sell the product.
  40. 3:26 And if you are a finance team or marketing, we want to help you with the research.
  41. 3:32 So each one of the disciplines, we want to help you to doing the job.
  42. 3:37 And together we have four big bets on the AI Work Platform.
  43. 3:40 We have Monday Sidekick, we're gonna talk about it mostly today.
  44. 3:44 Monday Vibe, if you want to build your own software.
  45. 3:47 Monday Agent, if you want to create your own agent into the platform.
  46. 3:50 And if you want some more deterministic flow, we have Monday Workflows.
  47. 3:55 But today we're gonna focus on Sidekick.
  48. 3:57 Sidekick is your intelligent AI personal assistant that understands your work, think and execute
  49. 4:04 with you, like a bird on your shoulder.
  50. 4:06 He knows you, he knows your business.

Chapters

  1. 0:00 From system of record to system of context
  2. 0:56 The gym answer: data without understanding
  3. 2:36 monday.com, Sidekick, and where work lives
  4. 4:31 Three reasons context is hard
  5. 7:19 The Monday world model
  6. 8:15 The data model and its two engines
  7. 10:21 Why the split mirrors the brain and lambda architecture
  8. 11:14 How it comes together, and the honest limits
  9. 13:22 Answering the question with Sidekick

Open at this second