Led · Seed · Jun 2026

Fika Jobs builds a video-first hiring platform where candidates connect a LinkedIn profile, complete roughly 10-minute AI-powered interviews and receive short video clips that form a live profile employers can browse. The company's AI generates personalized interview questions after reviewing a candidate's background and currently uses Google's Gemini models to power the interview agents. Rather than applying to individual roles, candidates maintain a discoverable profile that employers can revisit as opportunities arise. Fika positions its product as a way to surface communication skills and cultural fit early in hiring, and the founders say the approach may benefit early-career and non-traditional candidates. The platform has been tested by more than 50 companies and has over 100 companies on the waitlist. Fika plans to launch more broadly later this year, initially focusing on Sweden, and aims to grow its team to around 10 employees by year end. Its commercial model charges employers a 10% fee on a candidate's first-year salary after a successful hire while remaining free for job seekers.

Participated · Seed · Apr 2026

Founded in 2024 by Anders Hammarbäck and David Österdahl, Redpine builds a headless API platform that unlocks premium, non-public datasets by working with content owners in a compliant way. The platform combines proprietary retrieval and reranking technology with real-time data evaluation to improve the quality and reliability of AI outputs and ensure data providers are compensated. Redpine focuses on curated datasets across scientific and other high-value domains and cites applications in areas such as clinical guidelines, case law, physical research, financial markets data, and quality human-created news. The company plans to scale globally and expand its proprietary data partnerships while collaborating with AI labs and partners like AsedaSciences. Financially, it recently raised €6.8M in a seed round led by NordicNinja with participation from Luminar Ventures and node.vc to fund product development and hiring.

Participated · Series A · Mar 2026

Photoncycle is developing a distributed, solid-state hydrogen energy storage system that converts surplus summer solar electricity into hydrogen stored underground and later releases it as heat and power during winter. The technology is aimed at households and small businesses, closing the critical seasonal storage gap that short-duration batteries cannot solve. Photoncycle plans to offer the solution via a subscription bundle that includes solar panels, the storage unit, servicing, and access to energy-trading markets, reducing upfront costs for consumers. At full industrial scale, its proposed 1.4 TWh annual manufacturing facility could equip roughly 140,000 homes with 10 MWh of seasonal storage each. Initial commercial rollout is planned for Denmark and the Netherlands, with longer-term expansion opportunities identified in Japan and the United States. The company recently secured €15 million in Series A funding to finance these initiatives and the first phase of its industrialisation plan.

Led · Equity · Feb 2026

Berget AI is a Swedish startup developing a full-stack platform that allows developers to deploy open-source language models on infrastructure physically located in Sweden. By keeping data and models within national borders, the service helps enterprises and public-sector users comply with the EU AI Act’s transparency, traceability and data-governance rules. The company positions itself as an easy-to-use alternative to U.S. hyperscalers and complex on-premise setups, offering quick onboarding without the need to manage hardware. Its first product, launched shortly before the summer of 2025, has already attracted significant inbound interest from customers that prioritise data sovereignty. Co-founders Christian Landgren and Andreas Lundmark draw on backgrounds in digital entrepreneurship and AI consulting, respectively. With several new services in the pipeline, Berget AI plans to accelerate both product development and commercial expansion. The newly secured capital will be directed toward scaling the engineering team and ramping up sales operations.

Participated · Seed · Oct 2025

FirstQFM AB is a Stockholm-based quantum technology startup creating proprietary machine-learning foundation models aimed at enhancing quantum computer performance, scalability and availability. The company’s patent-pending models tackle device calibration and error correction, delivering measurable gains on Noisy Intermediate-Scale Quantum (NISQ) hardware. By partnering directly with quantum hardware manufacturers, FirstQFM seeks to shorten the path to fault-tolerant systems and make quantum capabilities accessible to industries such as chemistry, finance, life sciences and mobility. Early pilot projects on superconducting machines have already demonstrated significant performance improvements, and broader pilots across additional quantum modalities are planned. The founders, CEO Vish Ramakrishnan and CTO Isaiah Hull, lead a global network of partners and experts supporting the technology’s rollout. To date, the company has raised €1.2 million in pre-seed funding to finance ongoing R&D, pilot execution and commercialization efforts.