Inside Ramp, the $32B Company Where AI Agents Run Everything | Geoff Charles

If you're not using cloud code this year, no matter what your role is, you're probably underperforming compared to others on the company. PMs often

pride themselves on like the spec, the perfect spec. They have to understand

perfect spec. They have to understand that it's actually AI that's reading the spec now versus engineers. 50% of RAM's code is built by AI. And that's 50% up from 30% in December. It'll probably be 80% by March.

>> And this is not just like a front-end prototype right?

>> This is the real product, back end, front end, and I have a PR and I can just submit it to the engineer team. PMs

are shipping tons using Inspect. So are

designers, so are operators, so are like account managers and sales people are also getting activated. My job is to automate my job and our all our jobs is to automate our jobs.

All right, everyone. My guest today is Jeff uh CPO of RAMP and RAMP is one of the fastest growing companies ever and probably the most AI native company that I know outside of the big labs. So last

year uh Jeff and the team shipped over 500 features and hit over a billion dollars in revenue all with around 25 pfms. So yeah, really excited to talk to Jeff today and uh welcome Jeff.

>> Super excited to be here. Thanks for

having me Peter.

>> Awesome man. So um you know I've worked at a lot of big tech companies but like can you give us a quick overview of how ramp ships features like from idea to launch? Yeah, I I'll skip I'll skip the

launch? Yeah, I I'll skip I'll skip the basics and and just jump into the fact that it's a crazy time right now and uh the way that we are building has always

been around velocity and and the way that you move fast is by leveraging tools and AI is just an incredible accelerant to um everything that we do and you know I hope during this call

that you know I'll be able to share a few of the ways that we've we've leveraged AI to accelerate um to inspire folks and and and help amplify then the learnings I also expect that a lot of the things that we're going to talk

about today are going to be outdated, you know, even by the time that that you even share this this uh this recording.

So, I'm excited for it. But yeah, I mean the the product development process uh you know hasn't dramatically changed in terms of root principles, right? It's

about understanding customer painoint about identifying the right solution about uh building the solution and then testing and iterating and and I think um AI just lowered the cost of each of

these sections uh dramatically. you know

the cost of code is basically down to almost zero apart from the tokens and so uh PMs just need to uh be actually writing the specs for um the agents rather than the engineers themselves and

I think that's a that's a complete shift in terms of of um how we go about it.

>> Yeah. So so basically the PMs will make the will make the product first pretty much by themselves right or make the prototype at least and like get some validation before doing anything else.

>> Yeah. Yeah, I mean we we you know PMs often pride themselves on like the spec, the perfect spec and um they have to understand that it's actually AI that's reading the spec now versus engineers.

And so >> um the spec itself is is is basically the output of a prompt and then the output of the spec is the product. So um

at the end of the day it's just prompt to product back to prompt back to product. Um and yeah, we are essentially

product. Um and yeah, we are essentially um collaborating on an actual product itself and a prototype. Um I would even call it a prototype. it's actually a working product rather than um the

actual spec itself.

>> Yeah, I always suspect that engineers don't read my specs carefully. So like I always I always try to keep my specs to like less than two pages to begin with because you know no one wants to read the but like yeah the AI agent will actually thoroughly read re read it. So

so that's a good thing. Okay. So before

we even get to the spec though like first you have to like like you said you have to understand the customer understand problem and and like how how do you how do you guys work with AI to figure out what what to build or what the customer painoint is?

>> Yeah. Yeah. So the advantage that we have is that you know we have we have 50,000 plus customers on ramp and growing super super fast. We have over a million end users and so that gives us a ton of signal. We also have a ton of

people on on sales on support on account management. And so those are all touch

management. And so those are all touch points um that we can leverage to understand kind of what the problems are and what opportunities are and and what we should be focusing on. The the the question is around like how do you

actually sift through all this noise and that's where like a large language model is fantastic. So the first thing we

is fantastic. So the first thing we invested in is um what we call voice of the customer and uh typically was you know it was it was a person that we hired that that tried to you know do all this work uh themselves. Now, it's

basically an agent and and that agent is essentially able to sift through all our Gong recordings, all our Salesforce notes, um all inapp surveys, all support tickets, all inapp chats, um any email

that is being sent to um to account managers and essentially gather all that context as well as our Snowflake database and our analytics and help answer any question that that product managers have around their persona, the

pain points, their their workflows and the gaps of their products. Um, so happy to jump into that, but that's that's a huge thing that we've invested in.

>> Yeah. Do do you want to do you want to uh are you able to give a live demo of of that or like show us how that works?

>> Let me share one version of that. So,

>> yeah, >> this is our voice versus the customer tool and um and you know, as you can see, you know, you can ask any question for on this on this on this bot and this bot will literally go through any type

of question. So, you know, for this

of question. So, you know, for this demo, I asked, you know, what's feedback that we people have on our procurement product, right? And uh you can you can

product, right? And uh you can you can see that the sources. So, do you want me to look through support tickets, chat logs, sales research, feature requests, etc.? I said, "Okay, let's just go

etc.? I said, "Okay, let's just go through support ticket and chat logs."

It literally went through 90 days of support tickets and chat logs and identified the actual key uh topics that we needed to focus on as well as links

to the underlying uh assets for me to double click into. So, you know, purchase order management, approval, flow routing, uh chat, you know, uh chat understanding with ramp assist, exports,

currency constraints. I mean this is

currency constraints. I mean this is like this was done in you know from 38 to 40 so about eight minutes um and something that would have taken eight days for a human to actually do across the entire volume.

>> Yeah I mean this is basically like kind of prioritizes your road map for you like has number of support tickets and and and everything.

>> It at least helps you identify with a ton of context the problems that your your customers are facing with and enables you to go deeper. So then you can you know it's it's essentially a conversation right? Imagine you have

conversation right? Imagine you have this like full-blown analyst. How do you continue prompting the analyst to go deeper? So now it's like, okay, I want

deeper? So now it's like, okay, I want to go super deep on this specific problem case. Bring customer quotes.

problem case. Bring customer quotes.

Bring me some like log rocket sessions.

Bring me um uh you know customer IDs that I can go and research to. Create a

email that I can go that I can use, draft in my Gmail account for me to actually automatically send to set customer to book meetings on my behalf.

All these things are basically prompts and and this agent actually has all the connectivity to be able to do these things. And I love how the user

things. And I love how the user interface is just like a Slack channel or or like I guess you can DM the agent too if if you want.

>> Yeah, 100%. We've seen like Slack being a great place to actually host these things because that's essentially what you would do with a human, right? You

would you would Slack your your product operator or Slack a team channel being like go do these things. So, it's a very natural way of of of uh of doing work and personalizing these these agents as essentially your co-workers.

>> Oh, wow. Okay. Okay.

>> This episode is brought to you by Granola. If you're in back-to-back

Granola. If you're in back-to-back meetings, you know how much work it is to take notes live and clean them up afterwards. That's why I love Granola,

afterwards. That's why I love Granola, the best AI meeting notes app in the market. Here's how I use it. Granola

market. Here's how I use it. Granola

automatically takes notes during a meeting, and I can add my own notes, too. After the meeting ends, I use a

too. After the meeting ends, I use a Granola recipe to extract clear takeaways and next steps in the exact format that I want. Then, I can just share her notes directly in Slack with my colleagues or even get Granola to

share her notes automatically. Honestly,

of all the AI apps that I use, Granola is the one that saves me the most time.

Try it now at granola.ai/peter

and use the code Peter to sign up and get three months free. That's

granola.ai/pater.

Now, back to our episode.

>> So, that's that's a qualitative piece that you show me. What about the the metrics and the data piece? Like, how do you like you just give people access to like pull data themselves or how how do you Yeah.

>> Yeah. So, so there's uh you know the space is moving very very quickly. So, six months ago, you

very quickly. So, six months ago, you know, we we we launched uh our own bot for for data analysis. Um I'll give a quick a quick quick view of this and this is now outdated. Um and I'll tell

you why.

>> Um so we launched uh what we call ramp research. So it's funny like before you

research. So it's funny like before you would you would ask a data analyst or you would try to do it yourself with you know looker. Hex is getting pretty good

know looker. Hex is getting pretty good at like creating your own prompts etc. But it was still, you know, fairly fairly a lot of work to like get an answer to a question, right? And now

it's like, >> hey, I have a question. Give me the answer. Um, so uh now we have ramp

answer. Um, so uh now we have ramp research that essentially I mean the use cases are insane and actually by making it easier for people to ask questions about data, you actually you actually increase the number of people who

actually ask questions about data and you actually become more dataentric uh as a company. So you know what's an example? Um, you know, let's say that

example? Um, you know, let's say that you have an an automated email campaign and you want to understand the performance. What's the open rate of

performance. What's the open rate of automated emails that customers sent?

Boom.

>> You know, ramp research understands our entire database and understands all the schemas. It understands what you're

schemas. It understands what you're trying to do and automatically like generates the the actual interpretation of of set results. And this is, you know, this is used, you know, so much and this was this was in two minutes,

right? Um, by by literally everyone. So

right? Um, by by literally everyone. So

sales people trying to find, you know, what are our customers in Milwaukee? um

support people trying to figure out like the the common use cases of of XYZ product uh marketers trying to figure out the performances of their campaigns etc. But I I shared that this was

outdated um in the sense that we now basically have moved to uh Snowflake CLI plus cloud plus skills. Um so

essentially we we we've now moved to to using you know cloud code. Um we we have our own database of of skills that we've developed. So um we have a data anal an

developed. So um we have a data anal an analyst skill that essentially fully understands our database and understands how we go about approaching a data analytics problem and the best practices

of that. And then we essentially can now

of that. And then we essentially can now prompt cloud to say hey um you know bu you know build me a full report of the performance of you know our procurement product and identify you know the top

reasons why people um opt in the top the top uh uh blockers in our funnel um and you know draft with me you know 10 different growth ideas that we could be running and and and Claude will actually

generate you know a full HTML report fully baked into our data that is directly actionable.

>> Yeah. Okay. So it's not just Q&A anymore. is actually doing work for you,

anymore. is actually doing work for you, right? That's why it's better than the

right? That's why it's better than the thing.

>> Yeah. I mean, at the end of the day, like it's funny like you you you know, you ask a question, but you have a goal, right? So sometimes you ask the question

right? So sometimes you ask the question and you get the result, but uh you should just tell AI what your goal is.

And you'll actually be surprised at the uh questions that the AI can actually ask themselves to get to the goal.

>> And and um have have you given the entire company access to clock or is it like just engineers or everybody can you use it?

>> Anyone can use it, right? Um and in fact like you know we'll get into this around like how you actually become more AIdriven as a company but um if you're not using clot code um this year um no

matter what your role is um you're probably underperforming compared to others on the company and so it's certainly not um a product for for engineers um it is absolutely a product for builders and you know we talk a lot

about cloud code right now like you know opus 4546 like big launches and and a big big movement in the last like three months I mean you just saw the anthropic $30 billion raise, but I I I expect the

the tools to continue evolving. Like,

you know, by the time we we meet next um you know, the next 90 days, it might actually be completely different, right?

Um and so um uh it's less about like forcing people to use one tool. It's

about um giving people full access to any tool they want to share openly what where people are using and then get people to adopt to get to the aha moment. And then um but yeah, we don't

moment. And then um but yeah, we don't want to be dogmatic, but we want to like radically empower and and also have like full visibility on what people are doing.

>> Yeah, let's talk about later, man.

Because I I I think so many companies still don't get this. They're like, oh, you know, like what's the cost of this?

What's the ROI of this? Why should I give some salesperson this thing? Just

don't get it, man. We'll talk about we'll talk about later. Yeah. So, let's

keep going. Let's keep going down the product development process. So now you have all this crazy uh feedback coming and the quant I mean people talk about product sense like it's some mythical thing but I think it's just like like how much product feedback are you

getting like how how much are you like embedded in like the the feedback loops and like the the data every single day and then you kind of just develop that as a second nature kind of thing you know so but so but after you have that

like how do you actually um you know you mentioned that you don't actually read specs anymore like how how do you define the solution you just make the make the product right off the or >> yeah so so there's the in there's the

problem identification then there's like the actual like writing of requirements um and we have you know our own clot skills for that right so so you know claude has has full access to our notion um notion has the full context and all

all our personas um and all the research we've done that's all automatically transcribed um and aggregated and then we have skills in in in cloud that is

like your your product spec skill um and we've prompt we've designed it so that um it's a conversation based approach.

So just like you have maybe like a a manager or a peer reviewer on your spec uh claude will actually like interact with you and ask you clarifying questions. So for example, you know

questions. So for example, you know what's the main goal here? What are the main here are the here are the trade-offs like should we trade out this or that? Like have you thought about

or that? Like have you thought about this? Like what is the intersection with

this? Like what is the intersection with that? It has all the context uh about

that? It has all the context uh about what we're trying to do because it has all the other projects that we're building and all that is is is in notion. And so it it helps you basically

notion. And so it it helps you basically refine and and and get to an end state.

But yes, we don't really talk about the spec itself. That's just a step in the

spec itself. That's just a step in the process. We talk about the actual uh

process. We talk about the actual uh product. And so happy to happy to share

product. And so happy to happy to share a little bit of like how how fast we move in terms of prototypes and and and show you kind of what we're talking about here.

>> Yeah. Yeah. Please show us the skill and and you know everything else. Yeah.

>> Yeah. So uh let's let's go through like you know an example of this skill um and then we'll talk we'll talk about the actual uh how we build. So this is an example of of like you know clot skill

like you know folks should be pretty well verssed in in this in this in this world. Yeah. So you know product shaping

world. Yeah. So you know product shaping um defining like the role right push for simplicity surface trade-offs surface questions um you know key key definition

of the problem um you know like looking up like uh you know all the data that we have access to um do the actual research uh look at different competitors

customer evidences it has links to all these different things um synthesize the completion help me shape this question so you know you know present the the synthesis ask 20 questions around you

know, the forcing decisions, um, all the different principles that we have, um, and then like related skills. So, like

this is like a skill that you that that we we load up. And this this was actually just built by by one of my PMs. Um, other PMs actually have their own skills. We're trying to figure out like

skills. We're trying to figure out like how we actually get to like one strong skill as part of like the the the evaluation process. But this is one of

evaluation process. But this is one of the examples I want to share.

>> So, you basically just have to So, you basically just be like, hey, I want to build like some expense tracking feature and this thing actually drives the conversation, right? you actually drive

conversation, right? you actually drive the conversation with with you. Yeah,

>> exactly. Okay.

>> And then let's go into like the build.

So so so you know 50% of RAM's code is is built um by Yeah. Um and and and that's 50% up from 30% uh in December.

It'll probably be 80% by March. And um

you know, it's it's not inconceivable for it to be like 90 to 100%.

>> Yeah. Like we've hit we've hit coding uh escape velocity and and and it's a brave new world out there. um we the the question becomes like how do you make it

easy for a non-builder uh to engage with you know code because it's it's obviously pretty intimidating.

So uh we invested a lot in in building um our own uh visual on top of of uh you know any any large language model um and

to radically accelerate like how builders can build and how even like PMS can build. I mean if if you uh if you

can build. I mean if if you uh if you have like infinite coders at your disposal you know you are actually the bottleneck and you actually need to like um start moving faster. So I'll give I'll give you an example. Let's say that

you actually want to uh uh you have a lot of feedback from customers saying hey I need more visibility on um what needs my attention um and I need to understand kind of what's overdue uh

what's on track and like what's upcoming. So I need to understand like

upcoming. So I need to understand like my accounts payable cash flow. Okay.

>> Yeah. Yeah. So all I need to do is I will go in and say, "Okay, please build me a report on top of this table that

has four metrics. Uh my overdue paint, my overdue bills, my upcoming bills, uh 0 to 30 days, 30 to 90 days, and the total amount outstanding that we'll need

to pay." Okay, so this is obviously a

to pay." Okay, so this is obviously a shitty spec. This is just for demo

shitty spec. This is just for demo purposes but the but inspect will go and it will actually implement this product.

Um and it will understand the task it needs to do. It'll actually plan it'll understand the codebase. Um you've

already directed it exactly to where you actually need um and it also has access to our design component library. Right?

So I don't need to teach it to design, you know, what a metric should look like, what a module should look like, what a click should look like. It it has all these components baked in. And so

it's actually able to just reuse a lot of our existing code to build this thing. And I actually did this yesterday

thing. And I actually did this yesterday for just this demo purposes. So

>> yeah, >> you know, here here's here's like where where it gets to. See if this is working.

Boom.

>> Wow.

>> So now now you have on top of the bills table your entire uh your entire metrics. uh what's past due, what's

metrics. uh what's past due, what's coming up, and the total to pay. And and

this took five minutes.

>> And and this is not just like a front-end prototype, right? This is like the real product or >> this is this is the real product, back end, front end. Um and I mean, a lot of this is front-end code though because I

I didn't need to create more endpoints like like we already have all these endpoints, but inspect is able to do uh both front end and back end. Dude,

that's because I I've been using like, you know, Google AI Studio and stuff just to make prototypes, but that that's just like pure front-end code. It's not

you can't actually push through prod.

But, uh, sounds like >> and and it doesn't have context on your codebase. It doesn't have you need to

codebase. It doesn't have you need to you need it doesn't look the same as your product. Um, uh, no, I mean here I

your product. Um, uh, no, I mean here I can I can literally now I can, uh, I I can go in and, you know, I have a PR and I can just submit it to the engineer

team and we have an automatic PR review processes where, you know, a double- digit percentage of our PRs are automatically approved. So PMS are

automatically approved. So PMS are shipping um tons uh using inspect and so are designers so are operators so are like

you know some some extent like account managers and and and sales people are al also getting activated um on this on this piece. So it's it's just a massive

this piece. So it's it's just a massive accelerant and and the number one users are also engineers um engineers using Inspect and and and here's the other crazy thing about about um this technology that I want to share. Um,

>> yeah.

>> So, you know, often times you you you have feedback. So, we we we we love

have feedback. So, we we we we love feedback. We obsess over like customer

feedback. We obsess over like customer feedback. We have, you know, tons of

feedback. We have, you know, tons of Slack channels where people are just constantly posting things, right?

>> Mhm.

>> It's very overwhelming once you have, you know, the number of customers that we have. Um, you know, this is an

we have. Um, you know, this is an example of of a uh a UX channel. Um, and

you know, basic thing, right? Hey, like

treasury is a product should probably be case sensitive, right?

>> Yeah. um add inspect in the web repro change the following sentence PR merged like like this is just one and this is a very easy thing

>> but like anything any question that you have say you're an engineer say where do I get started there's an escalation a problem like anytime there's an escalation on ramp uh you know AI takes a first step it understands exactly what

happened it understands where in the codebase it creates the actual PR And uh oftentimes it ships it. Same thing with uh with support tickets. Anytime there's

a support ticket that comes in where someone is confused uh we have inspect basically run through that and and and recommend changes and have the PR up and ready for like the PM or the product operator or even the engineer to review

and ship it like that. The speed at which we can move with these some of these things is like radical. And this

is >> kind of have the AI give the first first pass. Yeah. Yeah.

pass. Yeah. Yeah.

>> First pass everything. Yep.

>> Let me push back on a little bit. If

everyone in the company is shipping these PRs all day, how are you going to keep the product cohesive or like keep the quality bar high?

>> A lot of the the PRs themselves are like quality of life improvements. Um

>> uh we also, you know, within Inspect, we have an understanding of like complexity and so we we do have a process by which we review things based on like the sheer amount of complexity that it has. Um and

it does route to the right person based on you know whether this is like a big change on the product side or on the engineering side etc. Um but we haven't we haven't yet gone to a big problem. We

also have a pretty robust release process. So once the PR is merged like

process. So once the PR is merged like we will like slowly roll it out and um before any like major changes happen um on the product that goes to the rest of our customers we have an automated process by which you know I get involved

or the directors of product get involved as well.

>> Okay. Okay. So, so you have the typical like um first like everyone in the company plays away with it and then you know if if nothing breaks you get them beta users to play with with it and then you loyal to all users.

>> Yeah. Exactly. So we so we have we have doc fooding alpha is like your your your customers that that are um you know as part of your research group. Beta is

anyone that opted into the beta tier. We

have about you know 10% of our customer base that's in the in the beta tier. So

you can launch very very quickly to beta and then you track you track uh analytics on that. And then to go from from beta to Ga we basically uh you know require for for large large announcements like not not really like

this you know naming convention or anything like that. Um for any large feature uh that is material to to the customer we basically have a review process that's fully automated. So so

because everything is you know you know in in our databases we have another bot you know uh ramp releases that uh creates a ramp release report. It it it it pulls all the information of the

context. It pulls uh uh you know a

context. It pulls uh uh you know a preview of the actual product that we can use. Um it pulls from our our

can use. Um it pulls from our our snowflake databases the impact that this feature has had. Um it pulls from uh you know any slack channel a summary of all the work that was done. Um and it it

basically like you know synthesizes all the things that uh it also can create it can do work. So to do release you basically need you know a help center article well that it gets written automatically. you probably need like a

automatically. you probably need like a internal enablement of like what this feature is, how do you use it, why do we build it, it writes that automatically.

You can also post in Slack. Um, so yeah, that's a little bit of like how we how we speed up that process.

>> And when you review these like larger features, are you reviewing the actual product or you reviewing because you know like a lot of companies just like that the PM writes some sort of document, right? And then they go

document, right? And then they go through like multiple rounds of reviews and then and and then you approve it and then and then they finally go build a product. But like I don't I don't think

product. But like I don't I don't think that's how it works at RAM, right?

>> Yeah. I mean the the question is like what is what is my role in all of this now, right? And and and I think you know

now, right? And and and I think you know in the past my role was well you know I'm the best at like the craft or I'm the best at like understanding what customers want or I'm the best at you

know understanding the data and that's no longer true like you have a super intelligent like platform that you can leverage. So yes, like I I I will try my

leverage. So yes, like I I I will try my best to um look at all the all the customer feedback, make sure that this is actually meeting the customer feedback. I will look at the metrics and

feedback. I will look at the metrics and call on like this is not good enough or this is not big enough. That's

something that is fairly subjective. I

will go into the product and test it out and play with it and like really just hone in on on on like what's working, what's not working. But I think the higher level job for leaders now is

>> based on your feedback, what broke down in the in the process, >> right? So if if you caught a a poor user

>> right? So if if you caught a a poor user experience, >> Mhm.

>> what broke down, what prompt failed, what skill failed, what design system failed, because giving feedback to the person and so that they can just fix it,

that's that's that's a >> that's a one-time band-aid, right? What

you want to do is you want to figure out within the process what broke down and fix that process so that next time you never have that feedback again. Like a

classic example for me is like I've asked I I've told the team 10 times the the call to action >> needs to be above the fold.

>> That's like you want to you want to you want to just you know six years of of of AB testing you want to increase conversion. It's a big button that's

conversion. It's a big button that's above the fold. That's it. And and and I' I've said that like 10 times or maybe a hundred times but now it's part of our design uh create process which is a

fully automated uh process in and of itself. And so before before it gets to

itself. And so before before it gets to me uh you know within within our Figma prototypes uh th those those core concepts are actually fully integrated.

>> Okay. Got it. Okay. So you don't have to yeah you don't have to say the same thing over again. You can provide maybe like higher level feedback or something.

>> Yeah.

>> My job is to automate my job. That's and

our all our jobs is to automate our jobs. Um

jobs. Um >> uh we can talk about what happens next, but yeah.

>> And and how about the how about that cuz another thing that sucks up a lot of time is like this annual planning process of like oh I spent like a month to figure out what we're going to build for the year or like for the next three three years, right? Like h how do you

guys manage that process or is there even like a like how far do you how far do you guys look on the stuff?

>> Uh honestly 3 months.

>> Okay. We we we can only predict within 3 months now. And and by the way, like

months now. And and by the way, like within 3 months, you can do what you can do in 3 years now. So like 3 months is actually a really long time.

>> Yeah.

>> Um you know, planning planning for me is um I think there's actually like three main objectives to to to planning. One

is actually aligning on strategy, which is much much more important. Like what

problems are you focused on?

>> What are problems are you not focused on? And which customer segments are you

on? And which customer segments are you going after? And and how are you

going after? And and how are you thinking that we're going to win longterm? like what is the the end state

longterm? like what is the the end state for this thing? So, it's about trade-offs and I think like the conversation should really be about trade-offs. The second thing that that

trade-offs. The second thing that that planning is good for is just having some level of commitment from the teams, right? Um some level of accountability.

right? Um some level of accountability.

And the third is um to have some baseline for sales to know what's coming >> um for them when they when they when they when they talk to a customer and

the customer asks okay like this is great but you know um I I I I have a lot more needs when it comes to our international exposure and the sales team needs some basic assets and so

that's that's that's the third kind of pillar and that also is like fairly automated. So once the team kind of does

automated. So once the team kind of does you know their their backlog and and their plan uh in notion we have an automatic like uh process that creates uh you know one pagers and then it

creates slides and content for the sales organization within our own uh branding guidelines. Um and then the sales team

guidelines. Um and then the sales team can essentially just like look at a higher level view of our road map to be able to sell effectively against it.

>> Wow. Okay. And then you have all these like vision and like you know how we're going to win stuff. Obviously AI can read read it and if something changes you can just ask AI to update it. Is is

that is that how it works? I mean what what I ask AI to do is is to synthesize information like a lot of a lot of leadership time is about

helicoptering between the nitty-gritty problems and then the higher level like strategy and roadmap and and and making sure that every level of the organization understands information at the bottom and information at the top

like how you communicate to the CEO and the board is very different than how you communicate to the director is very different to how you communicate with the teams and that LLMs are incredibly good at because and so like the

translation layer, right, when I'm at an all hands meeting versus when I'm at a team meeting versus when I'm in a boardroom. Very very different. And so I

boardroom. Very very different. And so I I waste a lot less time on those things.

Got it. Got it. Okay, great. Let's get

to the key question then. I mean, you just mentioned that your job is to automate your job. I'm sure all your PMs feel the same way. And so so what's going to happen to the PM function? Do

do you think >> do you think it's game over or like uh >> Yeah. What's going on? You know, it's

>> Yeah. What's going on? You know, it's it's funny like um I was I was surprised by like once you automate code like the a lot of people

concluded that PMs are it's like over for PMs and and I and I was I thought to myself it's over for the engineer for most engineers. So maybe it's like a lot of engineers who are like I'm going

to be a PM now because >> uh the engineering function is is is um has changed a lot. Now obviously there's there's a ton of value for engineers because um I think an engineer now is

managing hundreds of thousands of agents um and they can actually scale their impact.

>> Um but let's go back to the PM the PM role like a lot of what a lot there's a lot of bad PMs out there or badly trained PMs. I think that the way we've trained PMs in the past has been really

really bad and we've trained them on stakeholder management. We've trained

stakeholder management. We've trained them on prioritization. We've trained

them on communication. We've trained

them on frameworks. And those are all outdated because code is free. And so

like all that matters now is are we going the right direction? How fast can we go? And how do we remove bottl necks

we go? And how do we remove bottl necks and how do we build a system by which like we can accelerate and and and to do that I think PMs need to really rethink their skills. So like a lot of PMs join

their skills. So like a lot of PMs join product management because it's a safe job. You know they might not be good

job. You know they might not be good enough at the engineering task. They

might not be good enough at the design task but like they're really good at you know the the the consultant. I'm an

ex-conultant. Like that's why I joined the the function. Like I understand that the customer I can communicate to engineers and I can like I can I can really I can I can I can somewhat facilitate decision- making. The the

downside is that if you're a riskadverse PM, you're not going to change your way.

So I still see, you know, very high performing PMs who are who don't get it, who haven't yet adopted these these core skills, who haven't changed the way that they're working because it's worked for them so so far in their career. They've

been successful because of it. That is

the biggest danger that I'm seeing. So

um uh I I think that the the role of the PM is going to shift and I think it's going to shift in two directions. Uh PMs

are going to become much more builders, right? Because code is free. So just

right? Because code is free. So just

like I showed like uh a product, right, that I that I that I basically built in in in five minutes. Um it's going to require then like the iteration from the product very very quickly. And so I

think I think the craft and the building is going to be like really really essential versus the spec. like you no longer have to write the spec anymore.

You need to actually like be in the product itself.

>> Now an engineer, a great product engineer can do that and a great product designer can do that. The other path for uh product is um is the business side.

So you know what engineers are and and designers often lack is an understanding of the context in which the business operates and what actually matters and um and how we're going to win long term.

So they're really really good at maybe they're really good at building really good products and so give them that and then the product team should be focused on like okay but now that we have this really good product like how are we competing how are we positioning how are

we distributing how are we monetizing how are we actually using this to win and drive enterprise value and and and I think that you know even looking at open anthropic like the the the it's a it's a

it's it's a decision of strategy >> um they have different strategies and and that's actually where where the PM should be should be really really focused is is the underlying way that we're going to win and and playing that

GM mindset. Uh because because they're

GM mindset. Uh because because they're you're going to have a ton of builders that can that can build great products that can iterate on customer feedback that have all the context. You've built

that system. So now focus on like what actually no one can do which is to make sure that the product you're building is going to have insane amount of value in the market and insane amount of money for your business.

>> And like a lot of PMs are just like stuck like like you mentioned they're stuck in like cross functional alignment meetings all day like back toback. So

like and I I think it's like a company culture kind of thing too, right? Like

do do you do you make sure your pre PMs actually have time to build or is it sort of like do they have to get alignment from 10 people to ship anything? No, it doesn't seem like it

anything? No, it doesn't seem like it doesn't seem like that that's the case.

Yeah.

>> No, I mean we we've we've designed the organization so that we we do not have committees and we do not have signoffs.

Um you just need to prove that you've added value and then you can go go for the races. I I will say that like it's

the races. I I will say that like it's it's it's actually really really important for PMs to carve out time to uh build. And I say this not just PMs,

uh build. And I say this not just PMs, but like managers.

>> Um I I think that it's it's a really tough time to be a manager right now because you're managing a team whose skill set needs to change and and you might not actually have that skill set.

So I I I think that right now like going back to IC mode is is paramount.

>> Um and and I've done this for myself where I say like, "Hey guys, like I'm going to be in way less meetings. I'm

going to be way less one-on- ones and I'm going to be like I'm just going to be adopting AI tools. I'm going to be building and vibe coding and understanding what's working, what's not working so that I can be become more educated because it's it's just this is

just the beginning. I mean I mean the sheer amount of of changes that happened over the even the last like 3 months is is is profound and I think um if you're stuck in meetings you're not going to you're not going to you're not going to be effective. So definitely creating

be effective. So definitely creating space for for work and and honestly, you know, that's also where nights and weekends come in, which is like this is the year that like you need to you need to really prioritize learning and growth

because no one's going to do that for you. Um so yeah, uh uh it's uh it's

you. Um so yeah, uh uh it's uh it's going to be a wild ride >> and and if if doing the old way like your company's going to die basically, right? If it's do the the waterfall and

right? If it's do the the waterfall and all this kind of stuff, it's not going to survive. Yeah. Let's skip to uh

to survive. Yeah. Let's skip to uh talking about um companies that are watching this. They want to become AI

watching this. They want to become AI native like ramp like how you guys operate like how how do they go about like uh doing like you know building systems and that kind of stuff. Yeah.

>> So there isn't like one right way but I'll I'll share kind of what what we've done. Um

done. Um >> um and we've kind of like built built a framework around this. So so um we think about like being being AI proficient in like multiple levels. Okay. the the

bottom level is like people who sometimes use chat GBT, right? We'll

call them like the L0. Okay. The the

level one is like people who've built their custom GBTs, maybe they've built a notion agent, maybe they've built uh they've used like cloud code to like do some of these things, etc. Level level two is people who are actually like

fairly proficient. They they have been

fairly proficient. They they have been able to build an app that um that automates part of their job. uh they

have been able to commit uh code or feedback to other people's work. And

then level three is like the fundamental like systems builders. Okay. And our job is to get everyone in the organization up the ladder. And the way we do that is

as follows. The people who are still in

as follows. The people who are still in L0, they they will most likely not be at the company because the fact is like you can you can you can tell them as much as possible. If you're not a self-starter

possible. If you're not a self-starter and you don't have that growth mindset, like it's going to be very very hard to train to train you out. So, so um that's L0. The the L1's need to get L2s, L2s

L0. The the L1's need to get L2s, L2s into L3s. And L3 is like basically like

into L3s. And L3 is like basically like influence and the rest of the organization. And the way we do that is

organization. And the way we do that is um we have uh a lot of public channels around people sharing uh what they've built. We've made it really really easy

built. We've made it really really easy for anyone to adopt these things. So

we've removed any constraints around access, around tokens, around budgets.

We've uh we have like uh the setup of those tools are are are extremely well done. So you have access to all the

done. So you have access to all the different MCPs, you have access to all the different skills. We even have like an internal repository of skills that people are deploying to. You can pull from those. Um and then we have you know

from those. Um and then we have you know a lot of culture around you know in all hands around like showcasing non-builders doing things you know our finance team building their own treasury

management system our legal team you know doing contract reviews our marketing teams automating like website creation um to get people inspired and then we have office hours that that

people can join to uh to ask any questions to get set up. We have um designated experts that people can just ping and like their entire job is to get to evangelize, to get you set up, to get

you comfortable, to get you going. Um

those are like some of the the principles there. And then we we and

principles there. And then we we and then the other piece is just like you know hiring and performance management.

So on the hiring front, we we now have an absolute requirement for anyone who joins the company to be uh somewhat proficient for these tools. Um there's

just absolutely no excuses. And in the interview process, we'll have basically a dedicated session for this where like they will either I mean for the product team, I literally have a session where you you're going to build you're going to build a product like you're going to

show me a product that you've built and you tell me exactly how you built it, how you built it, how it works like it is a full-blown prototype.

>> Um and then we also track usage of AI across the company. So uh you know we have uh we vibe coded it this this this product even within the team where we

can see everyone in the company and their full usage of tokens across notion AI chat GPT uh cloud code cloud coworker um our inspect tools are or any of the

internal apps and we can see kind of like who is actually pushing the bar to amplify and who's not and who we need to contribute on. Do you worry about like

contribute on. Do you worry about like this cost stuff running out of control or like the ROI is so clear that there's no just just give everyone access, let them do it. Yeah.

>> I mean, I I I haven't done the ROI around like if you let's say you have a person who's who's has $100,000 salary, how many how many tokens should this person use? Um, and there's debates

person use? Um, and there's debates right now around you know you know uh >> uh productivity versus just like noise and you don't actually need these

things. I think right now we need to

things. I think right now we need to invest the budget for people to discover and if we are not as efficient in that spend that's okay that's our competitive advantage that's why we raised money

that's why we have a a pretty good war chest >> but um I can safely say that you know we pay our employees a lot of money and it and the the token consumption per

employee is not even close to double digits and I and I think it I think it's not unreasonable to think that it should be higher than your salary because like if if you have agents that that are able

to do 10 times more work than you, then why would you not pay them twice as much as you?

>> And so I think that's like the way that we should be really framing it. Um but

but yeah, I would say like we're not really worried about costs. Um we're

worried we're mainly worried around we have like the next x months or x years where um AI has not yet fully one-shotted a ramp platform um and and

we need to use that to our competitive advantage to to move as fast as possible.

>> Yeah. And I feel like a lot of the internal tools that you show me are also really great for ramp customers, right?

You can just like, you know, make that available for ramp customers.

>> 100%.

>> Okay. La last question, man. So, so if I'm a you PM or builder, like uh how should I think about my career DD these days? Like um you know like the old

days? Like um you know like the old climb the l ladder to VP or whatever like is that still going to work or how should I think about being employable still?

>> I would say um I think that the where you should be optimizing is not management. It is being the best builder

management. It is being the best builder in the world. I would say that management is probably dead. Um, there's always going

probably dead. Um, there's always going to be value in someone giving you feedback and coaching and and and and being your advocate and and being a team leader, but now is not the time to build

that skill set. Now is the time to like be very very proficient in this new technology and to um radically improve the the the way that you use it. Um, and

so I would say for for for all the PMs out there, um, you know, get really embedded in these tools. And that's why engineers are so good at at at you know understanding what AI is capable of

because they live and breathe it like that they're that's the first knowledge work that has been you know mostly automated with with with coding agents but it's coming for everyone else I mean it's going to come from PMs it's going to come for designers it's going to come

from for any any white collar uh job and so I would say just get very very proficient um and using these tools and um ultimately like the the your career is about impact and right now the impact

that you can have is to to um you ship great products faster and move more metrics for for customers into the business. And so um you know create a

business. And so um you know create a lot of space to learn these things um and and have the beginner's mindset, the humility to understand that the way you're doing things uh is not the best

way. Um and I think my job as a leader

way. Um and I think my job as a leader is is just to get people to to get to that aha moment. And even even my brightest PMs, I I had to sit down with them and say like we're going to go through this workflow together. what

what what have you done today and I will show you a new way of doing it and once once you get that aha moment that like red pill >> there's no coming back like like you were like I oh I get it now and it'll

also make you a better builder because your the software you're building if you're in B2B and even in B2C >> it is going to look radically different than than what exists today I mean fundamentally software is dead it's all

going to be like co-workers and if you haven't used co-workers in your own job you don't understand how like that actually might look like your product a lot more than you think, right? So,

>> yeah, you don't have to process it.

>> That's exactly right. Like like the ramp itself is going to look much more like a finest coworker than it does, you know, tables and charts and workflows.

>> Yeah. I find like um it's it's all like, you know, I've been using open call.

It's all like CLIs and like, you know, there's like no one wants to touch buttons anymore. It's just like let me

buttons anymore. It's just like let me let me talk to my co-orker and let you know, get him to do stuff for me. That's

basically it. Yeah. Cool. and and and and how do you build one of those great co-workers? Domain level expertise is

co-workers? Domain level expertise is another one. Like I think that in the

another one. Like I think that in the past it was like I'm going to talk to customers. I'm going to kind of

customers. I'm going to kind of understand the requirements and kind of build a product for them to do their job right?

>> But if you're doing the job of your customers so that they can do other things, you need to you need to actually be an expert or you need to build a system by which you can ingest that expertise. Right? So accounting like you

expertise. Right? So accounting like you can build an accounting workflow where they have to go and code things. But if

you're actually going to code on behalf of the accountant, you need to you need to deeply understand the philosophy or be able to extract that knowledge like how do you download, you know, CPA and all the best practices and actually bake

that into your product. So it's a very different way of thinking where fundamentally like a login in your product in the future um I think is going to be a failure, right? And I

think that's also how we think about it.

We we track the amount of time you spend in ramp and how we can actually reduce that time as much as possible which is by the way the opposite of how many PMs are trained. the the the Facebook and

are trained. the the the Facebook and Netflix, right? The the fangs of the world that

right? The the fangs of the world that are that are mainly advertising businesses like it is the opposite and I think um there's going to be reckoning for sure, but also a very exciting time.

I mean, you know, I think it's very very scary and a lot of people are alarmist and everyone should be paranoid, >> but man, it's an amazing time to be builder right now and especially a product manager where you have taste and

vision. The time it takes to go from

vision. The time it takes to go from your taste and vision to a product is shorter than ever. And I think it's a really really really exciting time to be a builder here.

>> Yeah. And I think uh another theme you mentioned is just like setting up systems to dedicate all the work to a AI, right? So you can focus on stuff they actually enjoy doing. Like

that's a key part of it. So yeah. All

right, Jeeoff. Well, I mean, thanks for being inspiration, man. Like I I I I think uh hopefully well hopefully every company can learn how to operate like rap.

Yeah.

>> We're just getting started. There's also

a lot of things that, you know, we're not doing well that other companies are doing super super well. I think, you know, part of me going on this this talk is not to to share that we've that we have that we have all of it figured out.

Most of the things that you saw here are things that we built in the last months.

>> Um, so excited to keep the conversation going, excited to continue learning and uh, you know, really really a privilege to to to be here today. Thanks a lot for for having me.

>> Yeah. SF.