• DeepScout attracted EUR 200K from MITON, with an option for an additional EUR 400K investment
• Serial entrepreneur Jan Mittner leveraged AI to automate market monitoring and optimize pricing for e-commerces
• DeepScout plans to enhance its platform, expand internationally, and attract early pilot testers

This January, Czech e-commerce B2B SaaS AI startup DeepScout announced having raised EUR 200K of Seed investment from the Prague-based startup studio MITON. As per the conditions, the investor reserves the right to invest EUR 400K more.

Origins and Beginnings

Jan Mittner, DeepScout’s founder, has over 20 years of experience in software development, primarily in web applications, with the last 10 years dedicated specifically to machine learning and AI. Since 2010, he’s focused exclusively on developing and operating his own projects (for example, Jízdomat and Proudly) rather than custom software development.

The serial founder himself particularly notes one success among these—an e-commerce store specializing in FPV drones, currently one of the largest of its kind in Europe. This was the experience that eventually led directly to the idea behind DeepScout: using ML and AI to automate comprehensive market monitoring, providing actionable insights for pricing and product assortment, enabling businesses to significantly improve their performance and simplify their expansion into new markets.

MITON’s partner Tomáš Hodboď tells ITKeyMedia that the journey leading to the investment in DeepScout took about a year and a half. Back in spring 2023, Jan Mittner had several different ideas in mind, and the concept initially described as a ‘co-pilot for e-shops’ or a ‘SimilarWeb for e-commerce’ appeared the most promising. The founder and the startup studio started testing the initial concept directly in the market—including MITON’s portfolio companies such as Knihobot and Bonami.

Essence and Development

‘For e-commerce businesses, growing competition makes it critical to know exactly what to sell and at what price to stay competitive. DeepScout helps businesses not only with pricing decisions but also provides insights on which products will deliver the highest return, ensuring that every euro invested contributes effectively to growth. For shoppers, DeepScout makes sure online stores offer products people actually want, at prices they’re comfortable paying. Ultimately, this creates a better shopping experience for everyone involved,’ Mr Mittner explains.

According to him, DeepScout’s AI technology was developed in-house, leveraging the team’s strong internal expertise in machine learning and AI. The solution combines several types of models: some are proprietary, while others utilize proven open-source models, and for specific tasks, proprietary models accessible through APIs are integrated.

‘Jan single-handedly managed to develop a robust and launchable MVP within a year. Thanks to today’s AI-powered development tools, this was equivalent to approximately three people’s work. It’s exciting to witness how AI boosts human capabilities and how much can be achieved even with a minimal team—though it certainly helps that Jan is an exceptional programmer. After a year of collaboration, we confirmed that even this initial version provides substantial value to customers. This convinced us that raising investment to accelerate growth and further develop the product was the right step,’ Mr Hodboď adds.

The initial prototype was completed within roughly six months, and since then DeepScout’s primary focus has been about significantly improving the accuracy and recall of the information and insights provided. In early 2024, the team already had a robust solution capable of handling millions of web pages and products, and in December 2024, DeepScout reached another major milestone: the technology evolved to the point where new websites could be added practically with a single click, and the entire training process was fully automated while maintaining high-quality results. Since February 2025, DeepScout has also been capable of overcoming highly advanced anti-scraping protections.

First Paying Customers and Use Cases

The startup can boast about paying customers since the summer of 2024, mostly domestic. The team is exploring the market and learning from real-world customers to identify areas where the solution already performs strongly, and others where it still has room for improvement.

‘DeepScout handles clearly differentiated products very well—particularly those with defined names, even if they slightly differ across various online stores, and consistent images.

It’s somewhat more challenging in sectors like fashion, where products often come in many variations such as colors and sizes, and are typically displayed on different models, making identification more complex. However, we’re making significant progress here and already have our first paying customers in this area,’ Mr Mittner shares.

DeepScout’s AI also successfully manages private-label products. This means that if an e-commerce store offers a private brand, DeepScout can effectively identify and suggest similar competing products. Although this requires manual verification of the matches, the number of private-label products usually isn’t very large, so the platform’s carefully optimized matching interface makes these checks quick and efficient.

Further Plans and Ambition

The recent EUR 200K investment will accelerate DeepScount’s ongoing enhancement of its product-matching capabilities, assessment of various use cases, and exploration of ways to obtain relevant data from any primary source cheaper and faster. This will naturally require attracting more talent, so the team welcomes everybody who feels like they can contribute, to reach out.

‘Personally, I really enjoy this investment style—dedicating over a year to a promising new application led by a talented team capable of quick iteration, product-market-fit exploration, and differentiation. In fact, given today’s rapid emergence of AI-first applications, I believe we’ll increasingly adopt this approach. It significantly improves the likelihood that ideas initially appearing as mere GPT wrappers can evolve into sustainable companies,’ Mr Hodboď comments.

Additionally, Mr Mittner’s team plans to start reaching out to customers beyond the Czech market soon, seeing how the platform has been designed to handle data from any country, in any language, and in any currency from the very beginning. As such, the company also welcomes early international pilot testers.

Without a doubt, AI-powered market and competition intelligence is essential for present day’s e-commerce businesses to stay ahead in an increasingly competitive landscape. By analyzing vast amounts of real-time data, DeepScout AI helps retailers optimize pricing, predict trends, and understand consumer behavior with unmatched precision. As e-commerce continues to evolve, leveraging AI-driven insights are poised to become the key to smarter decision-making, higher profitability, and sustained growth.