This week, we’re back with another installment in our Paths to Product-Market Fit series, this time with Linear CEO and co-founder Karri Saarinen.

Linear’s Path to Product-Market Fit — Quality and Craft > Speed and Scale

The metrics founders use to measure product-market fit — sales conversion rate, burn multiple, ARR — didn't matter to Karri Saarinen.

He's prioritized a non-numerical KPI since day one of building Linear: quality. Finnish by birth and a product designer by trade, he’s long balked at Silicon Valley’s obsession with scale and speed, which, he thinks, often comes at the expense of craft.

While a design lead at Coinbase and Airbnb, he grew frustrated with the product and engineering team’s project management software. “When I first started using it, I thought, ‘Why is it so messy and complicated?’” he says.

His Finnish friends Jori Lallo and Tuomas Artman, engineers at Coinbase and Uber, used the same tool at their jobs and shared Saarinen’s frustration. Over beers one night, the trio decided to team up to build a better one, figuring there must be a market of IC product builders just like them craving a faster and more elegant tool.

But before rushing to start a company, they let the idea simmer for a while, even running some lightweight user research with coworkers while still at their day jobs.

“Sometimes it's good to not build immediately. We didn’t want to commit right away and move really fast. We wanted to take some time to talk to people and form our thinking around this idea,” says Saarinen.

So when they set out to build the prototype over a year later, Saarinen says this incubation period made the actual work of designing the product very intuitive. ”Because we did so much pre-work and pre-thinking, a lot of the architecture was basically built into the product by the time we started,” he says.

Fast forward to today, and Linear now has plenty of numbers to show for it. Earlier this summer, Linear announced an $82M Series C, bringing its valuation to $1.25B, with a 15,000+ customer roster that includes the likes of OpenAI, Ramp and Vercel.

On The Review, Saarinen walks us through Linear’s methodical six-year build, detailing how the founding team…

• Built a highly opinionated prototype that met the co-founders’ high standards
• Launched in private beta and handpicked early users to join
• Stayed focused on the day-one ICP of IC product builders even with subsequent product launches
• Designed the sales process to be an extension of the product

Saarinen credits Linear’s success with this craft-first ethos, which hasn’t wavered even at its size today. “Quality is our first principle. Every other metric and decision flows from that,” he says.

Thanks, as always, for reading and sharing!

-The Review Editors

This week, we’re detailing exactly how Carta built an agent to solve one of its most pressing accounting problems.

The Dynamic Context Problem: How Carta’s Internal AI Agents Save Thousands of Hours of Back-and-Forth Work

Jayant Tikmani, Director of Machine Learning at Carta, finds the best problems for AI to solve based on this principle: deliver a differentiated customer experience and enhance service quality rather than reduce cost.

He found a worthy problem in the company’s fund administration business.

This service-based org of 600 people helps 2,500 customers handle complex accounting, reporting and compliance work. There’s a wide set of tasks this team can do for clients, so when a problem arises, it takes a lot of time to figure out what’s happening, what to do and how to fix it. Scaling is bottlenecked by context-gathering, which CPO Vrushali Paunikar says is an industrial engineering challenge.

So Tikmani worked with domain experts in the fund admin business and engineers to pinpoint workflows where agents could understand a task, build relevant context and suggest or complete next steps.

The result was the creation of an agent to handle cash reconciliation — turning an 11-minute task into one that could be completed in seconds. There are 20,000-25,000 of these tasks per month, which saved Carta’s internal team over 3,500 hours per month. Here’s how he and the team did it:

• Use a focused, low-investment PoC to assess risk, feasibility and answer key questions. Could the agent retrieve the right context from an internal tool via an API or database query? How should they structure and present the agent-generated diagnosis and recommendations to internal teams? Did users actually find the output useful and accurate enough to act on?
• Give the agent the right context for specific tasks and tools. Domain experts diagrammed the full workflow in Lucidchart, which acted as the source code for the agent’s system prompt. With that input, they created a set of gold-standard prompting techniques relying heavily on in-context learning: provide examples, the expected output structure and task framing.
• Lightweight design meant to evolve through iteration. Prototype and get a product in the hands of users to move as quickly as possible. To do this, they decoupled product performance (which is based on model behavior) from UX and workflow integration. They could evolve the agent’s reasoning and output quality without needing to constantly rebuild the product surface or its backend systems.
• Evaluating on one metric: helpfulness to users. Internal experts drove the agent’s feedback process. They provided a helpfulness rating via the product’s interface, compared normal task resolution speed to that of the agent and tracked reduction in back-and-forth between teams by monitoring escalations and handoffs.

This essay is a detailed look into how Carta identified a problem, set the right parameters for solving it and actually built the tool to do it. It’s an excellent example of AI in production driving meaningful business impact.

Thanks, as always, for reading and sharing,

-The Review Editors

Today, we’re excited to launch our new publication focused on how companies are using AI in production to drive real results. 

From Memo to Movement: Shopify’s Cultural Adoption of AI

We all know AI will transform every aspect of how we live and work. But the gap between that future state and our current reality still feels like quite the chasm to cross. 

As science fiction writer William Gibson once said, “The future is already here — it’s just not evenly distributed.” 

Most companies are navigating the “memos and demos” phase. Ambitious plans and mandates to start using AI are boldly declared in an ever-growing collection of CEO memos (initiated by Shopify co-founder and CEO, Tobi Lütke). Board decks trumpet “AI-first” roadmaps. Flashy features are center stage in the demoware that’s rocketing around social media. 

Yet scalable implementations and real results remain murky. 

That’s why today, we’re launching a brand new publication to close this gap: Applied Intelligence. 

For more than 12 years now, The First Round Review has published tactical playbooks from builders. Our obsession has always been with ground truth over grand theory, on mechanics of company building over market maps. Applied Intelligence extends that mission — to share how builders are using AI at their companies and the impact they’re seeing. 

Our inaugural essay is with Shopify’s VP & Head of Engineering, Farhan Thawar. Think of it as the sequel to Tobi Lütke’s famous memo, where we’ll explore the non-obvious insights, tactics and workflows Shopify used to bring the ambitious memo to life: 

• The surprising tools built by non-technical teammates when Shopify gave everyone access to expensive models
• The LLM proxy, which houses all internal agents and MCPs — allowing employees to build the workflows they need 
• The mentality behind hiring 1,000 interns to foster a “beginner’s mindset” and help teams solve problems faster 
• How they’re “context engineering humans” by identifying the high-impact workflows to inject AI 

You can read the essay here. And if you want to learn more about Applied Intelligence, go here. 

Thanks, as always, for reading and sharing!

This week, we’re looking at how founders can build out the initial version of their customer success orgs, from identifying the right people to hire, to the best early metrics to track.

The Founder’s Guide to Building a V1 of Customer Success

At some point, founders can no longer moonlight as customer success leaders.

The moment you realize it’s time to turn off support notifications can come in different forms — maybe something breaks and you’re drowning in tickets, or maybe too many high-profile customer requests pull focus away from other responsibilities.

Knowing when it’s time to build the CS function is one thing. Knowing how to build it is another.

Stephanie Berner, Atlassian’s SVP of Customer Success, is here to give you a blueprint. Prior to Atlassian, she led customer success for LinkedIn’s Sales Solutions arm, was the Senior Director of Customer Success Management at Box and gained early startup experience at Medallia, all the while learning the ins and outs of delivering strong customer experiences.

“Customer success is all about getting your product into your customer's hands as fast as possible,” she says. “If founders and product builders are spending disproportionate amounts of their time solving customer experience problems instead of driving the mission forward, it’s time to formalize your customer success operations.”

In this exclusive interview, Berner breaks down the crucial components of building an early customer success org, with specific zero-to-one tactics for founders standing up CS from scratch. Here’s what you’ll learn:

• Whether your first hire should be someone technical or an industry specialist
• Interview questions to ask that will evaluate drive, self-awareness and customer empathy (which are the most important traits of first hires)
• Why your CS leader should report to the CEO instead of the CRO or CPO
• Target ratios for the number of CSMs on SMB, mid-market and enterprise accounts
• Why customer health and renewal rate are the most important early metrics to track (and why NPS and CSAT can wait)

This is an invaluable, tactical approach for founders who are just starting to build their customer success functions.

Thanks, as always, for reading and sharing!

-The Review Editors

Recommended resources:

-How to calibrate your gut for hiring

-How to build a “board of brains” with AI