Game studios run on content. Thousands of strings of dialogue, item descriptions, quest text, UI labels, all of it needing translation into a dozen languages and version control across teams that ship updates weekly. For years that work lived in a swamp of spreadsheets, shared drives, and email threads where the latest version was whatever the most recent reply claimed it was.

A Swedish company has spent four years draining that swamp. Now it's raising money to put AI agents in charge of the cleanup.

Gridly, the Helsingborg-based content operations platform for game and software developers, has raised 1.3 million euros, around 14 million Swedish kronor, in a new round. The capital goes primarily toward weaving agentic AI into the platform. The round also brings in a name that carries real weight in the gaming world: Jörgen Larsson, founder and former CEO of the listed gaming group Stillfront, who joins as an investor and board nominee.

Gridly didn't start as a standalone idea. It spun out in 2022 from Localize Direct, a company that handled translation and localization for game developers. The founders, Christoffer Nilsson and Mattias Wennerholm, had watched studios wrestle with the same problem over and over. Content scattered everywhere, no single source of truth, and localization treated as an afterthought bolted on at the end of development.

Their answer was a platform that organizes, plans, and tracks digital content in one place. No more hunting through file versions. No more emailing the German translator to ask which build her strings are in. The core promise is boring in the best way: everyone working from the same data, all the time.

"Our platform already works very well for our customers as it is. Now, we're focusing on integrating AI, particularly agentic AI," CEO Anna Albinsson told Swedish outlet Rapidus. The pitch is that this next-generation AI will automate complex content workflows, pushing Gridly from a place where you store content to a system that actively moves it through production.

It's the same shift you're seeing across software right now. The tools that used to be passive repositories want to become active participants. For a localization platform, that might mean an agent that detects new dialogue, routes it for translation, flags inconsistencies, and updates every downstream build without a human chasing the thread.

On paper, 1.3 million euros is a modest round, with most of it coming from existing investors. The headline isn't the size. It's Jörgen Larsson.

Larsson built Stillfront into one of the largest mobile gaming groups in the Nordics, a public company assembled through a long string of studio acquisitions. He understands the operational guts of running games at scale, across many titles and many languages, better than almost any investor Gridly could have landed. His joining as a board nominee brings exactly the kind of domain expertise that a content-ops company selling into game studios needs in the room.

Gaming is Gridly's most important vertical, and Larsson is gaming royalty. An angel like that doesn't just write a check. He opens doors to the studios that make up the company's entire target market, and he can pressure-test the product roadmap against decades of hard-won experience. For an early-stage company, that's worth more than another million in the bank.

The existing investor base reads like a who's who of Swedish early-stage gaming and regional funds. Nordic Game Ventures, Teknoseed, Innovum, Subvenio Invest, and Rendered VC, which led a previous round, all returned. When insiders re-up, it usually means the metrics behind the curtain are pointing the right way.

Gridly is wagering that content operations is one of the cleaner fits for agentic AI in the whole software stack. Think about why. The work is high-volume, rule-bound, and endlessly repetitive. Detect a change, classify it, route it, validate it, ship it. That's a loop, and loops are exactly what agents are good at.

The risk is the same one every AI-content company faces. Translation and localization carry nuance, cultural context, and brand voice that a careless agent can flatten into nonsense. Gridly's edge is that it sits on the structured data layer, the organized content with all its relationships intact, rather than trying to reason over a pile of loose files. Structure is what makes automation safe. An agent that knows exactly which string belongs to which build in which language is far less likely to make the kind of mistake that ships a broken translation to a million players.

Gridly calls itself a content-ops platform, and the gaming focus is the wedge, not the ceiling. Any software product that ships in multiple languages faces the same versioning chaos studios do. SaaS apps, e-commerce platforms, enterprise tools. The localization and translation management market runs into the billions of dollars globally, and it's growing as more companies go international by default. Gridly's structured-data approach travels well beyond games, which is part of why insiders keep backing it alongside operators like the new Nordic VC funds chasing globally ambitious teams.

The company has signaled global ambitions from early on, and the AI investment is part of that. Automating the grunt work of content operations lowers the cost of supporting more languages, which lowers the barrier for Gridly's customers to expand. A platform that makes going global cheaper sells itself to anyone with international plans.

Worth being clear-eyed here. Agentic AI in production content pipelines is still early, and the failure modes are real. An agent that misclassifies a string, or pushes a half-baked translation into a live build, creates exactly the kind of mess Gridly built its reputation cleaning up. Trust is fragile in this category. One embarrassing localization bug shipped to players and a studio starts wondering whether the automation is worth the risk.

Gridly's defense is its data layer and its incumbency. Customers already trust it as the source of truth. Adding agents on top of a system people rely on is a very different proposition from a fresh startup asking studios to bet on unproven automation. Earn the trust first, then automate. That sequencing is the difference between agentic AI that sticks and agentic AI that gets switched off after the first incident.

The macro tailwind for Gridly is simple math. Games are getting bigger, releasing globally on day one, and shipping live-service updates for years after launch. Every one of those updates carries new content that needs translating, version-controlling, and pushing to a dozen storefronts in a dozen languages. The localization workload doesn't end at launch anymore. It runs for the life of the title.

That structural shift is why a tool like Gridly gets stickier the longer a studio uses it. A single-player game you localize once and forget. A live-service title you localize forever, and the studio that's wired its content pipeline through Gridly isn't going to unpick it mid-flight. Recurring content means recurring dependence, and recurring dependence is the foundation of durable software revenue.

Gridly isn't chasing headlines with a giant round or a frontier model. It's doing something less glamorous and arguably more durable: taking a product that already works, that customers already pay for, and making it sharper with AI where the automation genuinely helps.

That's the unsexy path to a real business. Land studios, embed deep into their production pipeline, become the system they can't rip out without breaking everything. Add agentic features that save real hours, and the platform gets stickier with every release. Bring in an operator like Larsson who can accelerate all of it. None of that makes for a flashy demo. All of it compounds.

The flashiest AI companies are trying to replace the creative work. Gridly's just trying to make sure the dialogue ships in the right language, in the right build, every single time. One of those problems studios will pay for on day one.