Subvolts take Build AI-powered enterprise tools with a dual-agent system and focus on hiring. May 19, 2026 - Developers
- Learn about practical agent architecture and model selection for enterprise AI tools.
- Business
- Understand how to build "intelligent" enterprise software and strategic hiring in the AI era.
This is a solid interview with a founder building AI-native enterprise tools, discussing practical implementation and strategy.
Rebuilding IT From the Ground Up for the AI Age: Serval's Jake Stauch
Audio streams from Sequoia Capital. Your place is remembered on this device.
From Subvolts
Summary and analysis by Subvolts editorial, AI-assisted.
Jake Stauch, Serval founder and CEO, discusses building enterprise software for the AI era on Sequoia Capital's Training Data podcast. He explains Serval's dual-agent architecture, using OpenAI for user interaction and Anthropic for code generation. Stauch also shares his views on hiring strategies and why he believes foundation AI labs won't enter his market.
Imagine a smart assistant for businesses that can automatically create new tasks using plain English. This episode explains how one company builds such systems, using different AI brains for talking to people and for writing computer code. It's about making work easier and more efficient with AI.
This episode offers insights into building AI-native enterprise tools. Stauch's approach to "boring" software with "unlimited intelligence" is relevant as companies like Google Gemini and OpenAI explore enterprise applications. Listeners learn about practical AI implementation, agent architecture, and strategic hiring in a rapidly evolving field. It's for founders, developers, and business leaders navigating AI integration.
- HostSequoia Capital
- GuestJake Stauchfounder and CEO, Serval
- Guest: Jake Stauch, founder and CEO of Serval
- Show: Training Data (Sequoia Capital)
- Length: 38 minutes
- Serval Architecture: Splits work between an admin agent (code generation from natural language) and a help desk agent (acts through approved tools).
- Model Usage: OpenAI models for end-user interaction, Anthropic models for code generation.
- Hiring Strategy: "Fewer, better" hiring as a durable moat.
Jake Stauch, founder and CEO of Serval, is building a ServiceNow for the AI era. His most contrarian bet is that the product should look like boring old enterprise software, but with unlimited intelligence. Serval's architecture splits work between two agents: an admin agent that uses code generation to spin up workflows from natural language, and a help desk agent that can only act through the tools admins explicitly approve. Jake explains why his team uses OpenAI models for end-user interaction and Anthropic models for code generation, why new model releases sometimes have to be rolled back when prompt tuning breaks, and why he's not worried the foundation labs will come downmarket. He also makes the case for "fewer, better" hiring as the only durable moat in a world where products may need to be rebuilt every six months.
Hosted by Pat Grady, Sequoia Capital
- prompt tuning
- Jake Stauch
- Adjusting the specific instructions given to an AI to get better results.
- moat
- Jake Stauch
- A sustainable competitive advantage for a business.
1 h 08David Silver and Lasse Espeholt of Ineffable Intelligence: A Superlearner for the Age of Experience
WithDavid Silver, Lasse Espeholt
1 h 041 h 0552 minMaking Cities Awesome: Peregrine’s Nick Noone & Ben Rudolph
WithNick Noone, Ben Rudolph
Same guests, other shows
Questions people ask
- What is Serval building?
- Serval is building enterprise software for the AI era, designed to look like traditional software but with advanced AI capabilities.
- How does Serval use AI models?
- Serval uses OpenAI models for end-user interaction and Anthropic models for code generation within its system.
Page generated Oct 19, 2026. Summaries are Subvolts' own; the audio, artwork and notes belong to Sequoia Capital.