WTF Is the Context Layer? The Missing Infrastructure for Production Agents

Talk transcript

1. 0:00

   [upbeat music] Hi everyone, uh, my name is Prukalpa. I'm the founder of Atlan.

2. 0:17

   Um, and, uh, today I'm gonna talk about this thing where context is having its moment. Uh, and so my goal today is to talk about, like, WTF is the context layer?

3. 0:29

   Um, just before I start, and I promise this is the last time...

4. 0:35

   Um, I don't know if the clicker is working.

5. 0:57

   Atlan, we... It's, it's working. Yeah. Thank you. Um, the problem we solve is we say AI doesn't know your business. We've fixed that. We work with an incredible group of companies around the world, ranging from GitLab and Zoom and Discord and Affirm to large enterprises like Mastercard and General Motors.

6. 1:15

   Um, and about a year ago, uh, my co-founder and I went on stage and we said, uh, at the dawn of the internet era, Bill Gates had written this very famous blog post, and it said, "Content is king."

7. 1:30

   Um, and as we are at the dawn of the agentic era, context will be king. Um, since then, it feels like twenty twenty-six is the year of context. Context graphs, anyone?

8. 1:41

   Um, uh, you know, every, every two days you see some version of context, uh, popping up. And so what is going on? Um,

9. 1:51

   I believe the answer to this kind of is in this reality distortion field that we live in. Uh, I live here in the [REDACTED:location]. Every day or two I have conversations with people which kind of go like, "How far are we from AGI?"

10. 2:05

    And we have a debate, and we're like, "Well, one year, three years," so on. Uh, there is no doubt that the models are getting exponentially smarter by the day.

11. 2:13

    Uh, two years ago, they couldn't pass the bar. Today, if they were to take the bar, it was-- they're at the top one percent of test scorers. On the other hand, they're not exponentially more useful by any benchmark.

12. 2:24

    Uh, one out of five, you know, AI use cases actually make it to production. Um, you know, fifty-six percent of CEOs say that there's zero financial benefit from AI today.

13. 2:37

    So what's going on? I believe hidden in plain sight is actually, um, how performance is measured in the human world. Uh, cognitive intelligence doesn't really determine real-world effectiveness. Uh, in fact, only ten percent of job performance variance is explained by IQ.

14. 2:54

    Like, just think about it. Would you say your smartest, um, you know, teammate who scored the highest on the SATs is also your best teammate? Or would you say, "No, it's the person who works the most and takes the most feedback and learns the fastest"?

15. 3:11

    In the real world, we care about performance, and performance is outcomes that you deliver in the real world. And performance is a function of two things. It's a function of intelligence, which is cognitive horsepower.

16. 3:22

    That's what the model benchmarks measure every day. Uh, but it's also a function of context. This is what they say in the human world as learning on the job, right?

17. 3:31

    Knowledge and skills and expertise that you learn over time.

18. 3:35

    And in the last decade, uh, we have compounded on one of those parameters. Uh, intelligence has thousand X'd in the last decade. Just in the last six months, we have two X'd on that axis.

19. 3:48

    On the other hand, context, the situated knowledge of your business, that's barely moved. We've moved some data to the cloud, uh, but that's about it. It's otherwise logged in dashboards and Slack threads, and, uh, the head of that analyst who might be leaving next week.

20. 4:05

    Um, and so the question ahead of us, and I really believe this is the next frontier, is how do we help AI build context about our business? Um, and every time I'm faced with a question about how do we help AI do this, I always like to go back and understand how did we help humans do this?

21. 4:25

    Uh, so I'm gonna take you into the life of, you know, a, a exemplar employee, Maya. Uh, let's say she's a data analyst at McContext Burgers 'cause I thought I was going to be creative, and I'm not very creative.

22. 4:39

    Um, and, you know, let's say she's that analyst that everybody, you know, pings in your company, uh, right? She's the person that everybody sends a message to every morning when they're trying to solve a problem.

23. 4:50

    So let's say this morning, uh, there's a franchisee owner who sends her a message and says, "Why is my drive-through time up this week? Why is this metric up this week?"

24. 5:00

    Sounds like a really simple question, um, but it's actually a really complicated question to ask.

25. 5:07

    Just to answer this one very simple question, Maya first needs to know, uh, what is drive-through time, uh, and who's asking. Is it finance or is it, you know, my ops team?

26. 5:19

    And it might mean different things. Uh, but not just that. What does this week mean? Is the cutoff period Monday to Sunday? Is it Pacific Time? Is it Eastern Time?

27. 5:30

    Uh, that's knowledge. Like, that's facts. That's the map of the business. Um, but not just that. Uh, there's expertise, uh, right? There's, um, you know, a diagnostic playbook. What, what does a great analyst do?

28. 5:43

    They know that, you know, quarter three is a season-seasonal quarter because of weather patterns, and they know to go check if the reason there's a spike is because of seasonality.

29. 5:54

    They also know that the company launched a product, uh, just that previous quarter, and so they know to check if that's why the root cause analysis failed. Uh, this is expertise and skills that people pick up over time as they learn on the job.

30. 6:08

    Uh, and then there's norms, right? Um, there's, you know, persona scoping. Who's asking the question? How do I answer this question? Um-

31. 6:16

    And Maya, she's one of those, like, cool people. She nails it. She sends an answer, not just with the answer, but with the why and the root cause, and she finds the reason for it.

32. 6:27

    How did Maya learn to do this? Um, she just joined the company a year ago. Um, first Maya, you know, has f- like, she joined, and she got some training, like all of us do, but that's not where any of us learn, right, in our companies.

33. 6:42

    How do we learn? We learn because you shadow, like, the best teammate, and then you see why they're doing something, and then you learn from that, and then you make a mistake.

34. 6:52

    Who here has learned more from a mistake than anything else? Right? You make a mistake, and then you learn. Uh, you- your manager gives you feedback, and you learn not to do that again.

35. 7:05

    You deal with an edge case, and then you learn from that. That's how all of us humans learn at work.

36. 7:12

    And so then the question is: How do you help build the agentic Maya? Uh, and now I wanna walk you through our experiments and learnings as we've built this at Atlan.

37. 7:23

    Um, era one, and this was roughly about 18 months ago now, um, we, uh, started on the, the track of bootstrapping agents. Um, and the way we went about it was, and we started this with our customer experience team, uh, and we did this jobs to be done analysis map, right?

38. 7:42

    And so we said, "Hey, if you are someone on our customer experience team, what are all the things that you do on a day-to-day basis?" And then we made some hypothesis.

39. 7:51

    We, we said, you know, for example, one part of the job is documentation and meeting prep. Uh, we said, "Well, AI could probably do that job pretty well." Uh, and so we built a scaling factor.

40. 8:02

    So on the other hand, relationship management is something that our customer experience team does, and we said, "Hmm, that doesn't sound like something AI is going to be able to do anytime soon."

41. 8:11

    And so we built a scaling factor. And then we basically started bootstrapping these individual agents that were, like, built for that specific topic. Uh, our team got creative, so we had Hermione, who is our health intelligence lead, and then we had, you know, Money Penny, who was our financial risk analyst, and we just made that particular agent

42. 8:35

    really good at doing that one thing. Um, and that worked for some time. Um,

43. 8:42

    but then we realized there were some challenges with this approach. The first, context engineering. Uh, we got to the point by middle of last year where building an agent was really easy.

44. 8:52

    It took, like, five minutes. Uh, but giving it the business context that it took to actually get it to be accurate took forever. Uh, quality of the agent often depended, uh, on the quality of context engineering, and that led to a lot of weird lost trust cases with our stakeholders.

45. 9:11

    Uh, then as we started taking this into production, we started seeing that these agents basically were kind of like living on their own island. Uh, now imagine, for example, if you're in a human team, and your marketing changes positioning on your, you know, uh...

46. 9:26

    And then they come to the town hall, and they tell you that they changed positioning. And so then, you know, the SDR on your team or your sales development rep, they know that they should use that new positioning.

47. 9:36

    This is like the infrastructure that we've built for humans inside our organizations. Agents didn't have that. So our marketing team had these agents, and they started making changes to that, and then our SDR agent on our website was still pitching the old version.

48. 9:51

    Uh, we had no idea how any of these things were even connected, so we didn't even know how to, like, run this as, as a team of agents. Uh, when an agent gets something wrong, this is hard.

49. 10:02

    Uh, it was really hard to, like, trace back what happened. Was it the model? Was it the agent? Was it the context? Like, where, how do we even go back and fix this?

50. 10:10

    Um, and over time, we started dealing with, uh, context sprawl. Uh, we had... The, the, the hard part about this was agents all had their own memory systems to a certain extent, so they were learning.

51. 10:24

    They were all learning separately, and they were learning differently. Uh, it became very, very difficult very quickly to say, "Okay, what does the single version of truth here look like?"

52. 10:34

    Um, and then over time, we actually went through, in the last 12 months, we've gone through cycles of, at the agentic layer, about 12 months ago, we were using one of these no-code type builders, uh, called Relevance.

53. 10:47

    We went from there into Google ADK, then we tried Glean. Uh, start of this year we moved to Claude Code. Now we're kinda like 50/50 Claude and Codex. Uh, and every single time as these changes happened, uh, our context got trapped in each of these individual systems.

54. 11:04

    Um, so start of this year, as general purpose agents started to become a thing, we said, "What if there was a different approach with general purpose agents?"

55. 11:15

    Uh, again, going back to the human world, well, Maya, she's not an individual star. She's part of a team, right? And, you know, you talk about these dream teams, like Maya and someone who runs customer support and someone who launches ads.

56. 11:28

    These people work really well together, and often these dream teams are built on shared context, right? Uh, they have a shared language. Uh, they have a shared picture of what's true today.

57. 11:40

    They have shared playbooks. Uh, they have shared norms, who's allowed to make what decision. Uh, and then they learn together. I think this is the most important part of it.

58. 11:50

    They have compounding learning loops of what good looks like, uh, and they have shared memory that, you know, "Oh, we launched this thing last quarter, and it, like, was terrible, and we're not going to make that mistake again," right?

59. 12:01

    And so we said, "Is there a way to bring that into the way we think about AI in our companies?" And so the mental model we started working on was we said, "Okay, we have these teams of humans, and they're across the board, and can these people essentially start building domain skills?"

60. 12:19

    So each of them is responsible for a certain set of skills. All of this goes into this common one place, which is this one- Company brain of sorts, right?

61. 12:30

    I like to think of this as the context layer. Uh, and then this has a bunch of retrieval mechanisms, which then talks to the general purpose agent across the ecosystem.

62. 12:41

    So then we started a experiment. Uh, this is some version of what our marketing team ended up building. So you'll see on the left, those are all the systems that our marketing team uses.

63. 12:52

    So data systems, our social and community platforms, our ad platforms, our analytics platforms. Um, and then you'll see this agent block. Uh, we build this very specifically for, um, having openness.

64. 13:08

    So we had Claude Code and Cowork. We also had our own Claude that we deployed, which has, you know, essentially talks in our Slack channels. Um, and then we used some external products, like Qualified and Artisan.

65. 13:21

    Uh, in the middle is kind of this context layer that our team started building. So think of it as our best SEO person was building their SEO skill. Uh, our best competitive intel person was building the best competitive intel skill, and that kind of became this common repo that we were building into and pulling out from.

66. 13:44

    This sort of became our living brain. Over time, we realized there were some things that we needed in this brain, right? Uh, we realized we needed a data graph, like if, for example, our autonomous ads agent, we realized it needs to do analysis on a daily basis, so, like, which table should I go pull from?

67. 14:00

    Uh, we needed a library of skills. We also needed some other things, semantics, metrics, what is ARR, how do you measure that? Uh, what is a qualified lead in our company?

68. 14:11

    Uh, and org structure, entities, things like that.

69. 14:16

    Over the last six months, we ended up creating about three hundred skills and forty agents in this team, uh, which has been incredible. Uh, but then with this approach too, we realized that there were some challenges.

70. 14:29

    We realized that context kind of needs to be managed like code. Um, so some challenges, let's pick skills. Uh, dependency management became really complicated. So, for example, we have this competitive intelligence skill, and it learns from the market on what's changing in the market, and it improves.

71. 14:49

    Um, it feeds our category positioning skill, which then feeds our sales battle card skill. Uh, now each of these skills is learning and evolving, uh, but every time they learn and evolve, it breaks something downstream.

72. 15:03

    Uh, and these skills very quickly start getting outdated and start drifting. Uh, who owns skill quality became another thing, like who eventually owns the quality of this? Security and governance was a nightmare.

73. 15:16

    Uh, we had secrets hardcoded in .env files. Uh, it, it was people were downloading these public skill repos. This-- The whole thing was like a nightmare. Um, and then I talked about context portability across all these multi-agent systems.

74. 15:31

    I started this talk by saying WTF is a context layer. Uh, these are the problems that a context layer is meant to solve. Um, the question I like to ask is: What does the GitHub for context look like?

75. 15:44

    Um, few thoughts. Uh, company context needs lifecycle management, collaboration, and versioning, uh, just like code does. Uh,

76. 15:56

    you know, there's questions like what's local context, what's global context, how do I keep this updated, so on. Uh, some thoughts in this, can skills have a profile just like code does?

77. 16:08

    Uh, can that have a self-learning, learning loop that's baked into it? Uh, what does quality management look like? Can you have security and postures-- posture management associated with that?

78. 16:20

    That's really, like, the first step. Uh, I see this as, like, having something that has built-in versioning and quality and dependency management. So you should be able to say, "Hey, this thing impacts all these other things.

79. 16:31

    This is the approver. This is the maintainer. These are the contributors." How do you build, like, kind of human plus AI workspaces that these, that these skills, uh, are managed via?

80. 16:43

    Second thing, every AI interaction creates more context, and harnessing this, uh, is gold. Uh, uh, there's been, I know, a lot of talks about self-improving loops. Uh, we have found that with traces, deploying a specific harness that actually is specialized in being able to go and reverse construct from that.

81. 17:02

    So think of it as AI that's reading through all your traces and almost brings it back to your maintainer loop and says, "Approve, reject, approve, reject. Improve this over time."

82. 17:13

    Uh, that's the compounding learning loop. And the third, often a lot of people ask me this question, which is like, "How do I start? Because my business is re-really disparate, and I have all these, like, sixty systems, and how do I even start?"

83. 17:27

    One of the biggest learnings we've had is context is hidden in these-- in business systems. Uh, and across this context, quality can really compound. So, for example, if you're able to connect your Salesforce and your HubSpot to your data warehouse, to your application layer, and then you're able to reverse construct how these things are actually connected one

84. 17:46

    to another, context today gets lost in every one of those hops. But if you can reverse construct that and then deploy AI on top of it, we've seen incredible accuracy in being able to reverse construct the first version of your company brain.

85. 18:05

    So I'll end with this. Uh, the way I think about a context layer is it's a system that turns knowledge and expertise and norms that we talked about, that Maya knows, into a machine-usable context for AI systems.

86. 18:20

    Uh, at a very high level, the way I like to think of it is it looks like this. Uh, it continually is mining context from your business systems. It's feeding this into that one company brain.

87. 18:35

    It's harnessing this in skills and context development life cycles as your teams go and deploy these agents, and then it has a bunch of ways you can retrieve it.

88. 18:45

    So MCP, SQL, Vector retrieval, hybrid assembly, all these different ways that you retrieve it and pull back from traces and build this compounding learning loop. [sniffs]

89. 18:58

    Today, we're largely building agents by hard coding context. [laughs]

90. 19:07

    The scale of this problem, I truly believe, is underhyped because with scale, this can become really unsustainable, uh, and a little dangerous. Like all of us know this, this old joke, which is if you ask sales and finance the revenue number, you're going to get two different numbers.

91. 19:22

    Uh, we're fast approaching a moment of starting to deploy autonomous systems where the same thing is starting to happen.

92. 19:31

    So I'll end with one last thing. I started this presentation by saying context is king. Um, I'd like to end it by saying context is also IP. Something I think a lot about is in a world where you and your competitor have access to the same models and the same intelligence, what differentiates a company?

93. 19:52

    What differentiates a customer support agent at American Express versus Amazon? Uh, that's how you do business. That's what makes your company special. Uh, context is how we take and encode our culture and our norms into something that we will be proud of as we build autonomous frontier firms, um, and

94. 20:17

    that's all I had. Uh, you can find me [REDACTED:username] on Twitter, um, or write to me. We are actively working with folks on the frontier on going and shipping and building company brains, um, so if you'd like to talk to us, feel free to reach out.

95. 20:33

    Thank you. [audience applauding] [outro jingle]