# Decagon fine-tunes smaller models for specific customer tasks

Markdown copy of https://realalpha.app/s/decagon-fine-tunes-smaller-models-for-specific-customer/ on Real Alpha, the best minutes from long interviews with the people building AI. Cite and link the page, not this copy.

- Speaker: [Ashwin Sreenivas](https://realalpha.app/builders/ashwin-sreenivas/) · Ashwin Sreenivas is cofounder and CTO of Decagon. · Co-founded Decagon, the conversational AI platform valued at $4.5B, after building Helia, which Scale AI acquired.
- Host: [Nakul Mandan](https://realalpha.app/people/nakul-mandan/) · Host, Knuckle Up with Nakul · General partner at Audacious Ventures; his show goes deep with founder CEOs on how they build.
- Show: Knuckle Up with Nakul
- Episode: [The AI founder advice you won’t hear on X | Ashwin Sreenivas (Co-Founder at Decagon)](https://realalpha.app/v/_FmwqPzqkjY/), May 19, 2026
- Moment: 57:19 to 1:00:09 of the episode (3 min), playing on https://realalpha.app/s/decagon-fine-tunes-smaller-models-for-specific-customer/; the episode from there: https://realalpha.app/v/_FmwqPzqkjY/?t=3439

## Quote

> “you can get a model to do better at the few things that you care about in a smaller model, which is hence **faster and cheaper**.”
> — [Ashwin Sreenivas](https://realalpha.app/builders/ashwin-sreenivas/) at [58:15](https://realalpha.app/v/_FmwqPzqkjY/?t=3495) · Knuckle Up with Nakul

## Takeaways

1. Frontier model performance can vary across capabilities.
2. Smaller tuned models can be faster and cheaper for narrow tasks.
3. Build in-house only when it improves customer experience.

## About this segment

Decagon fine-tunes smaller models for specific jobs in customer interactions, where frontier models may excel unevenly. Ashwin Sreenivas says they build only when it improves customer experience; otherwise, they partner.

From “The AI founder advice you won’t hear on X | Ashwin Sreenivas (Co-Founder at Decagon)”, May 19, 2026. [Watch the full episode](https://realalpha.app/v/_FmwqPzqkjY/)

## Transcript

Transcribed by machine (Whisper) from the episode’s audio, so a word may be off here and there.

[57:26](https://realalpha.app/v/_FmwqPzqkjY/?t=3446) model? It's a great question. The thing that we found is that at the frontier of capability, model performance is actually pretty jagged, by which I mean, just because you have a model that advances a lot in one capability doesn't mean that it does that in every capability. And in fact, in certain cases, it might regress. Now, let me give you a specific example. For our use case, We use a whole constellation of models in each interaction with an end customer.

[57:55](https://realalpha.app/v/_FmwqPzqkjY/?t=3475) And we care about each of these models doing a few specific things. Now, I don't care about that model being good at coding. I don't care about that model being good at writing poetry. I care about a few very specific qualities. Now, if I use the frontier model for everything, in a way, I'm kind of wasting a lot of parameters and capabilities that I don't care about. And so in our experience, you can get a model to do better at the few things that you care about in a smaller model, which is hence faster and cheaper.

[58:32](https://realalpha.app/v/_FmwqPzqkjY/?t=3512) And so therefore, if we want to create the best customer experience, we should fine tune and build a lot of those models ourselves. Because we get everything better performance, cheaper and faster. I'm curious, how do you decide what to build in-house versus where to partner? So, you know, voice is clearly a core aspect of customer support. How do you decide we should build voice models in-house versus partner with existing providers?

[58:58](https://realalpha.app/v/_FmwqPzqkjY/?t=3538) Our general approach to everything here is what can give us the best customer experience the quickest, right? By which if we can get best in class from a partner, we will always do that. if we think we can do better ourselves because it is better for the end customer, then we build ourselves. I think this example of Deck Gone Labs was a great one where for the capabilities, for the kinds of models where we think, you know what?

[59:26](https://realalpha.app/v/_FmwqPzqkjY/?t=3566) A frontier model does do the best. We can't improve on it all that much if we do it in-house. Great. I would prefer to partner with someone, right? Because again, now we get to leverage their R&D spend and capabilities kind of automatically get better. Whereas in cases where I think we can build better and it is crucial to the end customer experience, those are the things that we build in-house. Like I'm not too sensitive to saying, oh, this is really important. So we must do it in-house ourselves. Everything is about where can we improve on the end customer experience.

[59:58](https://realalpha.app/v/_FmwqPzqkjY/?t=3598) In your most paranoid self, where do you and Jesse worry about, hey, if the frontier labs keep