vidtheque.

read-only demo

Videos ITMXwI6QL6A

How Forward Deployed Engineering is done at Ramp — Leo Mehr

index_state ready data_status ok

AI Engineer· published 2026-07-28· 0:14:04· en-US· indexed 2026-08-10 19:38

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 0:46, 1 of 1 keyframes kept
  5. Shot 4, 0:46 to 1:01, 1 of 1 keyframes kept
  6. Shot 5, 1:01 to 1:34, 1 of 1 keyframes kept
  7. Shot 6, 1:34 to 2:21, 1 of 1 keyframes kept
  8. Shot 7, 2:21 to 2:47, 1 of 1 keyframes kept
  9. Shot 8, 2:47 to 3:14, 0 of 1 keyframes kept
  10. Shot 9, 3:14 to 3:40, 1 of 1 keyframes kept
  11. Shot 10, 3:40 to 4:06, 0 of 1 keyframes kept
  12. Shot 11, 4:06 to 4:32, 0 of 1 keyframes kept
  13. Shot 12, 4:32 to 4:58, 0 of 1 keyframes kept
  14. Shot 13, 4:58 to 5:29, 1 of 1 keyframes kept
  15. Shot 14, 5:29 to 6:00, 0 of 1 keyframes kept
  16. Shot 15, 6:00 to 6:31, 0 of 1 keyframes kept
  17. Shot 16, 6:31 to 7:06, 1 of 1 keyframes kept
  18. Shot 17, 7:06 to 7:45, 0 of 1 keyframes kept
  19. Shot 18, 7:45 to 8:13, 1 of 1 keyframes kept
  20. Shot 19, 8:13 to 8:41, 0 of 1 keyframes kept
  21. Shot 20, 8:41 to 9:09, 0 of 1 keyframes kept
  22. Shot 21, 9:09 to 9:37, 0 of 1 keyframes kept
  23. Shot 22, 9:37 to 10:05, 0 of 1 keyframes kept
  24. Shot 23, 10:05 to 10:33, 0 of 1 keyframes kept
  25. Shot 24, 10:33 to 11:23, 0 of 1 keyframes kept
  26. Shot 25, 11:23 to 11:52, 0 of 1 keyframes kept
  27. Shot 26, 11:52 to 12:21, 0 of 1 keyframes kept
  28. Shot 27, 12:21 to 12:51, 0 of 1 keyframes kept
  29. Shot 28, 12:51 to 13:12, 1 of 1 keyframes kept
  30. Shot 29, 13:12 to 13:40, 0 of 1 keyframes kept
  31. Shot 30, 13:40 to 13:47, 1 of 1 keyframes kept
  32. Shot 31, 13:47 to 13:51, 0 of 1 keyframes kept
  33. Shot 32, 13:51 to 14:03, 0 of 1 keyframes kept
  34. Shot 33, 14:03 to 14:04, 1 of 1 keyframes kept

34 shot(s).

keyframes kept every frame deduplicated

What was stored

cues
136
whisperx 136
chunks
24
from 136 cues
keyframes
15
kept of 34 captured
frames with text
14
142 lines read
chapters
9
from the source metadata
keyframe bytes
3.5 MB
word timings on 136 cues

Provenance

Each pipeline stage, its state and the model that produced it
stage state model started took
fetch done — 2026-08-09 22:12 0s
stt done — 2026-08-09 05:47 14s
chunk done — 2026-08-09 05:47 0s
text_embed done — 2026-08-10 19:38 1s
keyframe done — 2026-08-09 05:47 1m 46s
ocr done — 2026-08-09 05:49 4s
frame_embed done — 2026-08-10 19:38 2s

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

  • 0:35 #3 done2 line(s)

    shot 3·sharpness 179.1

    1. AlEngineer0.98
    2. World's Fair1.00

  • 0:56 #4 done10 line(s)

    shot 4·sharpness 1825.6

    1. AlEngineer0.96
    2. World'sFair1.00
    3. About me0.97
    4. Director of Engineering at Ramp0.99
    5. Joined 2.5yrs ago when FDE was 2 people0.99
    6. PRESENTED BY0.97
    7. Run a 30-person org including FDE, Dev API, and Al Services0.99
    8. Microsoft1.00
    9. World's Fair0.94
    10. Engineering the future of Al1.00

  • 1:27 #5 done20 line(s)

    shot 5·sharpness 2088.4

    1. AlEngineer0.99
    2. SALES1.00
    3. World'sFair0.97
    4. ENGINEER1.00
    5. TECHNICAL1.00
    6. ACCOUNT1.00
    7. PRESENTED BY1.00
    8. MANAGER1.00
    9. Microsoft1.00
    10. SOLUTIONS1.00
    11. CONSULTANT1.00
    12. FORWARD1.00
    13. DEPLOYED0.98
    14. ENGINEER1.00
    15. t210.78
    16. 80.92
    17. th 1.2K0.95
    18. World's Fair0.98
    19. TRACK 8· JUNE 30, 20260.96
    20. Forward Deployed Engineering0.99

  • 1:44 #6 done11 line(s)

    shot 6·sharpness 1850.0

    1. AlEngineer0.97
    2. World'sFair1.00
    3. FDE at Ramp1.00
    4. XEng org0.98
    5. Help Ramp win upmarket0.98
    6. PRESENTED BY0.98
    7. Support core product for our enterprise customers1.00
    8. Microsoft1.00
    9. World's Fair0.97
    10. TRACK 8· JUNE 30, 20260.96
    11. Forward Deployed Engineering1.00

  • 2:42 #7 done6 line(s)

    shot 7·sharpness 1574.5

    1. AlEngineer0.98
    2. World'sFair1.00
    3. Always Be Scoping1.00
    4. World'sFair1.00
    5. TRACK 8• JUNE 30, 20260.96
    6. Forward Deployed Engineering0.99

  • 2:58 #8 skipped

    shot 8·duplicate of #7

  • 3:34 #9 done7 line(s)

    shot 9·sharpness 1439.3

    1. AlEngineer0.98
    2. World'sFair0.96
    3. SAP0.95
    4. S/4 HANA0.99
    5. World's Fair0.98
    6. TRACK 8· JUNE 30, 20260.96
    7. Forward Deployed Engineering0.98

  • 3:48 #10 skipped

    shot 10·duplicate of #9

  • 4:09 #11 skipped

    shot 11·duplicate of #9

  • 4:42 #12 skipped

    shot 12·duplicate of #9

  • 5:19 #13 done6 line(s)

    shot 13·sharpness 1329.6

    1. AlEngineer0.98
    2. World'sFair1.00
    3. iOS0.98
    4. World's Fair0.96
    5. TRACK 8· JUNE 30, 20260.96
    6. Forward Deployed Engineering0.99

  • 5:56 #14 skipped

    shot 14·duplicate of #13

  • 6:28 #15 skipped

    shot 15·duplicate of #13

  • 6:39 #16 done11 line(s)

    shot 16·sharpness 2018.0

    1. AlEngineer0.97
    2. World'sFair1.00
    3. PRESENTED BY1.00
    4. SCALINGWITH1.00
    5. MODELCAPABILITIES1.00
    6. Microsoft1.00
    7. FALLING BEHIND0.98
    8. GENIWDO0.85
    9. World's Fair0.97
    10. TRACK 8· JUNE 30, 20260.96
    11. Forward Deployed Engineering1.00

  • 7:37 #17 skipped

    shot 17·duplicate of #6

  • 7:59 #18 done30 line(s)

    shot 18·sharpness 1811.1

    1. Thread # fde-requests0.98
    2. 全三0.63
    3. AlEngineer0.98
    4. World'sFair1.00
    5. N1.00
    6. NotionAPP Yesterday at 11:50 AM0.98
    7. added inFDE Intake·Yesterday 11:50 AM0.99
    8. New submission-1.00
    9. Submitted by: @Greg0.97
    10. Deadline: Jul 310.97
    11. Request Type: Enhancement0.98
    12. Product Area(s): Cards,Expense Management0.99
    13. Economics:1.00
    14. Open in Notion0.97
    15. 3 replies0.98
    16. FDE Intake Agent APPYesterday at 11:52 AM0.98
    17. Hi – I'll help scope this. To get started:0.96
    18. Customer outcome0.99
    19. 1) Can you confirm success looks like: all dining charges embedded on hotel0.99
    20. folios reduce that specific day's per diem balance, and if that pushes the day0.99
    21. over $125, the last card transaction of that day is flagged so the approver0.99
    22. requests repayment? (Notion)1.00
    23. Current workflow and data available1.00
    24. reliably submit hotel folios with line-level itemization1.00
    25. and dates for ancillary charges (e.g., dining lines dated per day)? If not0.99
    26. always, what's the expected fallback when the folio isn't itemized or is0.99
    27. missing? (Notion)0.98
    28. World's Fair0.99
    29. TRACK 8·JUNE 30,20260.98
    30. Forward Deployed Engineering1.00

  • 8:17 #19 skipped

    shot 19·duplicate of #18

  • 8:50 #20 skipped

    shot 20·duplicate of #18

  • 9:26 #21 skipped

    shot 21·duplicate of #18

  • 9:59 #22 skipped

    shot 22·duplicate of #18

  • 10:19 #23 skipped

    shot 23·duplicate of #18

Transcript

136 cues· 2,168 words· 11,347 chars

  1. 0:13 awesome thank you guys awesome it's great to to meet everyone i mean i hope that after the talk you know if you want to come out and we can chat i would love to um cool so yeah today my goal is to share with you guys the two most important principles from what we learned doing fde at ramp so just yeah briefly a little bit about myself yeah i'm a director of engineering at ramp
  2. 0:36 I joined the company two and a half years ago when it was just, you know, FDE was just two engineers at the time.
  3. 0:44 And today, my org is about 30 engineers across four deployed developer API and our new AI services business.
  4. 0:56 So I know this is kind of a running theme, but like no one knows what FDE is.
  5. 1:00 So I'm just going to spend a moment on that.
  6. 1:06 So, yeah, I actually kind of like this meme.
  7. 1:09 To me, it's kind of funny.
  8. 1:12 But I actually think it's totally wrong.
  9. 1:14 I don't see this as the actual true form of what FDE is.
  10. 1:20 I don't see it as the final evolution or boss mode of technical go-to-market roles.
  11. 1:27 Now, this might be true at some companies, but at least at Ramp, it's a little bit different.
  12. 1:35 So FD at RAMP, we live within the engineering organization and our goal is to help RAMP win up market.
  13. 1:45 So with that in mind, what we do is we basically work on the core product and our new agentic features and make them work really well for our largest enterprise customers.
  14. 1:59 So that's just a little bit of intro context.
  15. 2:02 I want to dig in and today,
  16. 2:04 Like I said, there's just two things I'm gonna share with you.
  17. 2:06 Literally two things, very easy talk.
  18. 2:10 And these are the principles that I would say have really guided us and I would say probably the two most important things that we have.
  19. 2:16 Always be scoping and scale with tokens.
  20. 2:21 So let's get started with the first one on scoping.
  21. 2:25 So I would say there's this thing where like, many people think that as an FTE, your job is to just say yes to the customer.
  22. 2:34 But that's wrong.
  23. 2:36 If you were just to say yes, instead of beautiful Waymos that we have driving us around in San Francisco, you'd have something like this.
  24. 2:45 Horses with rockets strapped to their legs.
  25. 2:49 And the point is, you want to help the customer be successful, you want to try to figure out a way to say yes, but you actually want to deliver good software.
  26. 2:59 You need to build the right thing.
  27. 3:01 So you don't just endlessly say yes to people.
  28. 3:04 And I do want to share an example of something that I would say happens somewhat regularly in one form or another at Ramp.
  29. 3:15 So it's Friday night, and an enterprise sales rep comes to us with an urgent request that this super important strategic logo is only going to close if we build out an SAP S4 HANA integration.
  30. 3:33 I think that the default engineering reflex is like, shit, what are the SAP API docs, where do I find them, and how do I build this integration?
  31. 3:43 But what a well-trained FDE would do is pause for a second and say, okay, first of all, what's driving the urgency here?
  32. 3:53 One thing I've seen is I've seen sales reps who like go kind of crazy because it's like the end of the quarter and they're trying to hit their quota and close the deal and not because the customer is the one driving the urgency.
  33. 4:03 So that's like one example.
  34. 4:06 But
  35. 4:08 As an FD, you're asking tons of questions to gather context about what's important and what actually is the right thing to build.
  36. 4:16 And so you might ask, who's using this integration?
  37. 4:21 Have we exhausted all the different workarounds?
  38. 4:23 Is there something manual that we can do in the meantime?
  39. 4:25 Does the customer have technical resources?
  40. 4:27 Can they hit our API so that we don't have to build this thing?
  41. 4:32 But I'd say the most important thing that an FD does is also looks beyond this one request and looks at the other prospects that are coming down the pipeline and other customers to see if anyone else would benefit from this as well.
  42. 4:48 And the point is that by gaining all this context, you can do a better job of building the right thing.
  43. 4:57 So I want to share another story that was really, really painful for us in the early days.
  44. 5:02 We had this large enterprise customer, and they needed this reimbursement feature on mobile.
  45. 5:10 Unfortunately, our mobile team was totally swamped.
  46. 5:13 So we basically just had to roll up our sleeves as FDEs and just find out a way to get things done.
  47. 5:20 And we had two of the engineers on the team just learn how to do iOS and Android development.
  48. 5:26 And it was awesome.
  49. 5:27 We were super excited.
  50. 5:28 We're like, OK, we're going to ship this feature.

Chapters

  1. 0:00 Introduction: FDE principles at Ramp
  2. 2:10 Principle one: always be scoping
  3. 3:16 The Friday night SAP integration ask
  4. 5:04 When scoping goes wrong
  5. 6:33 Using agents for the FDE pipeline
  6. 7:57 Automating request intake into specs
  7. 9:14 Building the spec agent in Notion
  8. 11:16 The agent harness and evals
  9. 13:27 The future of FDE needs both

Open at this second