Inovagent is an AI and software consultancy. We help teams adopt AI in a way that fits how they already work - practical implementations that support people rather than replace them.
[ About ]
We work from Stockholm and Seattle, two of the world’s leading hubs for technology and innovation. Setting up an AI workflow is quick these days. Making it something a team relies on every day takes engineering, and that’s what we’ve been doing for years. We start by understanding the people behind the process, then build systems that fit how they work and keep working long after launch.
Where we work
Stockholm
Local time in Stockholm:
59.33° N, 18.07° E
Seattle
Local time in Seattle:
47.61° N, 122.33° W
[ Services ]
We work out what your team actually needs, then point you to the right tools or build it ourselves, and keep it running after launch.
[ 01 ]
We look at how your team works and where AI or software would genuinely help, then give you a clear report on what to use. If an existing product fits, we recommend it. If nothing does the job, we scope a custom implementation we can build for you.
[ 02 ]
We build AI systems and custom software, from integrations to full applications. You choose how it’s run: we keep the code and maintain it monthly, keeping models and integrations up to date, or the code is yours and we’re there for maintenance whenever you ask.
[ 03 ]
Every system we build is designed around a clear division of work between people and AI. AI handles the repetitive and the scalable; people keep control of decisions, quality, and the moments that matter.
[ Prototypes ]
We build working prototypes to prove out ideas before they become products. A few examples:
An assistant built to help a hotel's team respond to guest emails faster and avoid losing potential bookings. Connected to the hotel's inbox and website, it could check room availability and pricing and draft helpful replies to private enquiries. By design, it never sent booking confirmations or processed payments: once a guest decided to book, a member of staff took over. The system ran fully operational for one hotel before the prototype was retired.
A chatbot that answered questions about education law in Portugal, drawing only on the publicly available consolidated legislation published on the official Portuguese government website. The prototype worked end to end on the data it had. The limiting factor was the source material itself: Portuguese legislation is highly fragmented, and even the “consolidated” versions were missing laws and articles, which left gaps in context.
An open-source personal knowledge base. It captures newsletters and informative emails alongside YouTube transcripts and stores them in Notion. A daily automation summarizes new content and adds it to a “Brain Map” that can then be queried directly. Planned work includes a fully local setup, an interface for AI coding tools, and more ways to capture entries.
github.com/inovagent/knowledge-gpt (opens in a new tab)A cafe discovery app built from publicly available data. Our AI pipeline classifies and enriches cafe information so TheMug can start with cafes that make the cut, then help people find the ones they love.
jointhemug.com (opens in a new tab)[ Clients ]
Primary consulting partner
Inovagent is the primary consulting partner SIDO relies on to develop its systems and data collection.
Game stats, team voice chat and player health come together in one platform, with a single login for every SIDO app.
Players' voice chat is recorded and transcribed by AI, so coaches can see how the team communicated in every game.
Staff can ask an AI assistant about games, players and health, and it only answers with what they're allowed to see.
[ FAQ ]
Straight answers to what teams ask before working with us.
Systems that take repetitive work off your team. We’ve built assistants that draft replies to guest emails, pipelines that classify and enrich data, and AI that transcribes team calls and answers questions about a company’s own data. We also build software with no AI in it, from integrations to full applications. Not sure where AI fits yet? Working that out is the first step.
Yes, that’s usually the point. We build around your existing tools and workflows instead of asking your team to switch: past systems have connected to email inboxes, websites, Notion and internal data platforms. If a tool has an API or an export, we can usually work with it.
We agree with you up front what a system can access, and give it only the data it needs. The AI assistants we build only answer with what each person is allowed to see. As a Swedish company we work under the EU’s GDPR, and we’ll walk you through where your data is stored and processed before anything goes live.
Send us a message through the form below. We start with a conversation about your team and the work you want to change, then scope a use case together. From there we build and test it with you on your own data and connect it to your tools, with people in control where it matters. You don’t need technical knowledge: you know the workflow, and we handle the rest.
Most implementations take 2–6 weeks, depending on the scope and how many of your systems they connect to. We start with a working version on your real data, so you can try it before we build it out. You’ll get a firm timeline once we’ve scoped the work together, before any build starts.
There’s no fixed price list, because every project is scoped to how your team works. Cost depends mainly on what we build and how many systems it connects to. After a first conversation, we send a quote for the agreed scope, so you know the cost before any work begins.
Yes, in one of two ways. We can keep the code and maintain it monthly: updating models, fixing issues and keeping integrations working as your tools change. Or you own the code outright and we handle maintenance whenever you ask. New features are scoped separately either way.
[ Contact ]
For project enquiries and consulting, tell us what you want to build.