Funding & VC

Wrike founder's Hops raises $35 million as AI push speeds engineers 5.3x

Andrew Filev forced his engineers to review AI-generated code same-day. Merged pull requests rose 5.3x, and his new AI workspace Hops has $35 million in funding.

By Grace Kim

4 min read

‘You either take the wheel, or we’re going to run into the wall’: The founder who sold Wrike for $2.25 billion forced hi
‘You either take the wheel, or we’re going to run into the wall’: The founder who sold Wrike for $2.25 billion forced hiPexels / pixabay

What's News

  • Andrew Filev's new company Hops raised $35 million, roughly two-thirds of it his own money.
  • Merged pull requests rose from 798 (Jan–Mar) to 4,211 (Aug–Oct), a 5.3x jump; 6.3x by net lines of code.
  • Filev sold Wrike to Citrix for $2.25 billion in 2021.
  • Hops opened to the public Nov. 10 after more than 2,500 teams in early access.
  • MIT's NANDA initiative found 95% of studied organizations got zero return on generative AI investments.

Andrew Filev's engineers merged 4,211 pull requests between August and October, up from 798 in the first three months of the year — a 5.3x jump he credits to forcing his team to work inside what he calls a "human-AI factory."

Filev has standing to run that experiment. He founded Wrike, the collaborative work management platform, and sold it to Citrix for $2.25 billion in 2021. After that, he built the AI coding agent Zencoder under his company For Good AI. Now he's launching Hops, an AI workspace for whole teams that opened to the public on Nov. 10 after more than 2,500 teams tried it in early access.

Hops has $35 million in funding, Fortune exclusively revealed. About two-thirds is Filev's own money; the rest came from a handful of angel investors.

How did he speed the engineers up?

Agents churned out pull requests faster than the team had ever handled, with one rule: an engineer reviewed every one the same day. Filev didn't pretend the old workflow would survive.

"Imagine I put them in a car, I gave myself a pedal and I floored it," Filev told Fortune. "And I'm like, you either take the wheel, or we're going to run into the wall. It's your choice, but I was still flooring it."

He was blunt with staff: "I understand your way is not going to work. So don't come to me and say it's impossible."

By net lines of code, output rose 6.3x. His gut says the team is at least 10x faster on business value, on top of roughly 2x gained last year from coding agents. The rest came from making each engineer own the whole job — product decisions down to code — instead of waiting on product managers.

Even his best engineers resisted. "I deliberately introduced shock into the system," Filev said. It took months, and he still re-explains concepts "three, four times."

What is Hops, exactly?

Think Slack crossed with a shop floor for AI. It has chat, shared docs, tables and file storage, with AI agents working alongside people. It connects to roughly 2,000 other systems, from Google and Meta ad accounts to Stripe.

The pitch targets where AI work lives. Most employees use ChatGPT or Claude Code in private sessions, so one person's breakthroughs never reach the next desk. In Hops, what one person teaches the system, everybody gets.

Filev's data analyst wired up the ad accounts once; now Filev just asks Hops for last week's campaign insights. "I have no idea what happens behind that," he said. Employees also build their own apps inside it, cutting or canceling seats in Tableau, Jira, Notion, Slack and HubSpot — about $135,000 in subscription costs, by his team's count. He calls it an "AI operating system for your business."

Why does company-level productivity lag?

MIT's NANDA initiative found last year that 95% of organizations it studied got zero return from generative AI investments. Filev's explanation is unglamorous: individuals get faster, but handoffs stay slow. If a designer cuts a task from two hours to 15 minutes, "the whole cycle is still going to take two weeks," he said.

Asked whether big companies are dissolving from the inside while AI-native startups move faster, Filev didn't hedge. "Absolutely," he said. "It is actually faster to build this from scratch." At Wrike, he wanted a foundational product change for years. "When I left the company, that change still hasn't happened," he said.

What happens to the jobs?

Hiring changed. Filev stopped testing coding skills — agents write the code now — and tests architecture much harder, screening for "agency": ownership plus the nerve to try things. Engineering managers are stepping back to manage agents instead of people. He grades managers on people, project and process management; with agents, people management "don't matter" and "you can yell at your agents all day long." What's left is spotting failure modes and fixing the system. He coaches his "battlefield-promoted 'agent managers.'"

On the broader backlash, Filev, born in Russia, calls himself a "romantic capitalist" and a "stoic": countries and workers that ignore the technology will "get hit," and the free market will sort it out, "either in a brutal way" or otherwise. His answer to an Engels'-pause scenario — gains flowing to owners while wages stall — is steering productivity into growth, not just cuts, and retraining people. "If the employees keep working the old way, your company is not going to achieve new results," he said.

Next in line: the marketing team, getting machinery with bars set "so high that there's no way that they can do their work unless they use the new machinery." And the fuzzy blue Hops meeting agent, already in beta with select customers, goes generally available in December.

Source: fortune.com

Share this article:

More from Grace Kim

Grace Kim

Show full bio

Market editor covering people, launches and day-to-day practice in entrepreneurship, startups and venture funding at Business Bearings.

2,582 articles

Related articles

« Previous article