The Agent-First Developer Stack

The Agent-First Developer Stack

Our investment in Linear.

August 26, 2026

Summary

AI has turned individual developers into force multipliers, capable of shipping in hours what once took weeks. The teams and systems that turn this new abundance of code into real organizational throughput will define the next generation of product development. We believe Linear is one of the companies best positioned to own that opportunity. That’s why we’re proud to partner with Linear through its employee tender offer: backing not just the company, but the people who built it.

  • Founders: Karri Saarinen, Jori Lallo, Tuomas Artman
  • Location: San Francisco, CA
  • Industry: AI Infrastructure

The Opportunity

The bottleneck in software development is moving. As AI models generate more and more of the code itself, the constraint shifts from writing software to managing what gets written: deciding what to build, triaging what comes in, reviewing what agents produce, and keeping a growing volume of work legible to both humans and machines. This is a structural change, and it creates a massive opportunity for a system that can serve as the connective tissue between human intent and AI-driven execution.

That system needs a clean, opinionated foundation. Platforms built up over years of custom fields, inconsistent workflows, and manual processes accumulate a kind of complexity that humans can work around but that AI agents struggle to interpret. The winners in this next era won’t just be the tools that generate the most code; they’ll be the ones that give agents (and the people directing them) a clear, well-structured place to plan, delegate, and track outcomes at scale.

The Solution

Linear launched in 2019 as a beautifully crafted issue tracker for software teams. That same design discipline is now the foundation for something much bigger: a product system built around context and agents. Over the past year, Linear has expanded from a single planning tool into a five-part lifecycle spanning intake, planning, build, review, and monitoring, with AI agents woven into every stage rather than bolted on as an afterthought.

That architecture is showing up in how teams actually work. Coding agents are already active across a large share of Linear’s enterprise workspaces, picking up issues, writing code, and shipping results back into the system, with the volume of agent-completed work growing rapidly. Humans define intent in Linear’s structured planning layer and agents execute against it, with Linear serving as the shared memory that keeps both sides coordinated. That data-model discipline is proving to be a real advantage now that agents, not just humans, need to read and act on it.

Momentum is also expanding beyond engineering. What began as a developer tool is increasingly used by product, support, and operations teams as a single home for planning and tracking work across the organization, a natural extension of the same underlying system.

Why We’re Backing Linear

Salesforce Ventures invests in teams we believe will define entire categories, and Linear’s founders have that track record. CEO Karri Saarinen helped shape design systems at Airbnb and built the founding design language at Coinbase before co-founding Linear in 2019 alongside Jori Lallo, a former Coinbase engineer, and Tuomas Artman, whose engineering leadership at Uber included building the company’s mobile architecture. Together, they’ve built a company that has grown quickly while staying remarkably focused.

Our investment reflects a broader thesis: as AI agents take on more of the actual work of building software, the value shifts toward the systems that can plan, coordinate, and make sense of that work at scale. Linear’s opinionated data model and its early, deep investment in agent-native workflows put it in a strong position to be that system of record for the AI-built software era, first for engineering teams and increasingly for the broader organizations they sit within. We witnessed the effectiveness of Linear firsthand when Salesforce became a customer earlier this year, replacing a decades-old internal system that wasn’t purpose-built for the AI era. 

What’s Ahead

We see enormous room for this category to grow, as the unit of work in software organizations shifts from individual engineer output to a much larger and more collaborative pool of human-and-agent effort. Linear is well-positioned to capture that shift, and it’s one that’s critical to startups and enterprises alike. We’re proud to partner with Karri, Jori, Tuomas, and the entire Linear team as they build the operating layer for how software gets made, and we can’t wait to see what they do next.