PITCH group round 13

Web Summit Lisbon 2025 showcased a dynamic array of startups presenting innovative AI-driven solutions across various sectors. The event, hosted by Ms. Olena Bakalo, a marketing and business development expert, featured seven startups pitching their ventures to a panel of judges, all vying for a spot in the semi-finals.

(This article was generated with AI and it’s based on a AI-generated transcription of a real talk on stage. While we strive for accuracy, we encourage readers to verify important information.)

Matt Wicks, Co-CEO of MyContentScout, introduced an AI knowledge platform designed to combat the significant time knowledge workers spend searching for information. He highlighted that employees spend 1.8 hours daily, or 23% of their time, on this, costing a 1000-person enterprise approximately $2.5 million annually. MyContentScout aims to deliver information precisely when needed, in the required format.

The platform searches internal organizational data, including compliance, sales, HR, e-learning, and video content, without accessing the public internet. It offers AI-summarized answers linked to original sources, deep analysis of documents and videos, and the ability to push content to specific groups for onboarding or job roles. Mr. Wicks also noted its capability to repurpose content, such as converting documents to audio or creating digital flashcards from audio, saving considerable time.

MyContentScout generates revenue through per-user and usage-based models, with upsells like an AI coach and monitoring dashboards. Addressing competition from large language models like ChatGPT, Mr. Wicks emphasized its ease of implementation, rich content, and contained environment, avoiding the need for extensive enterprise upgrades. The company ensures GDPR compliance through rigorous processes, storing data in Europe and maintaining an auditable knowledge graph for content and user removal.

Dr. Finn Williams, Founder and CEO of Rumii, presented a healthtech platform tackling mental health challenges in young people. He stated that one in five young individuals live with a mental health condition, with 90% never disclosing their struggles. Rumii transforms smartphones into continuous behavioral health sensors, passively analyzing raw data to infer sleep, activity, social contact, and problematic smartphone use without requiring wearables.

Rumii delivers “compassionate nudges” and “clinically grounded conversations” as just-in-time adaptive interventions. Healthy actions are incentivized by building currency redeemable for apparel, encouraging engagement. The platform’s daily interactions build a robust dataset capable of predicting risk, improving treatments, and refining clinical trials. Rumii’s monetization strategy combines user subscriptions with licensing aggregated, anonymized insights to pharmaceutical companies and partnering with CROs for remote patient monitoring.

Dr. Williams highlighted the team’s extensive experience in clinical psychology, deep tech, and pharma. Rumii is recognized by Innovate UK, boasts an 80% retention rate, and has Apple App Store approval for its full sensor stack usage. The company is raising £1.6 million to enhance its AI, scale consumer growth, and secure paid pharma and CRO trials, aiming for £1 million in annual recurring revenue.

Francisco Ferreira, Co-founder of Manie, addressed the significant problem of energy overspending. He estimated that consumers overpay by $2 billion annually in Portugal, $60 billion in Europe, and $380 billion globally, primarily due to lack of knowledge, time, and fear of switching providers. Manie serves as an energy hub, centralizing all energy contracts, invoices, and switches.

Manie’s platform allows users to switch providers directly, integrating with the entire market. Its flagship “AutoSwitch” product automatically scans for the best offers and initiates switches based on user preferences. Since its launch 12 months ago, Manie has processed approximately 3000 invoices, completed 50,000 switches, and manages 16,000 contracts, delivering $12 million in savings. The company projects $1 million in revenue for its first year.

Manie operates on a subscription and commission model, where providers pay for leads, and customers pay for advanced tools. Mr. Ferreira explained that the platform thrives in free energy markets, such as Portugal, Spain, and parts of the US, where competition allows for significant savings. Future plans include a bidding platform for suppliers, one-bill payment for businesses, and AI-powered optimization for energy usage and storage, with global expansion underway.

Ole Shved, CEO of StackBob, presented an AI-powered solution for employee access management, a $2 billion problem for mid-sized organizations. He noted that IT admins waste over 500 hours annually on manual provisioning and reviews, leading to $100,000 in lost licenses and a constant risk of data breaches. Traditional solutions are costly, complex, and often fail to integrate with all applications.

StackBob utilizes cloud-based AI agents for browser automation, enabling integration with over 300,000 tools in minutes without requiring API or SSO upgrades. This automates onboarding, offboarding, password management, and identifies/removes redundant accounts. StackBob’s strategy involves direct sales and channel partnerships with larger identity vendors like Okta, helping them manage disconnected apps and expanding market reach. The company, founded in Ukraine in 2022, has 50 customers and is expanding into the US, UK, and Nordics.

Mahsa Yarahmadi, CEO of Nixi AI, introduced an ambient healthcare AI agent designed to alleviate the administrative burden on doctors. She highlighted that doctors spend 33% of their time on admin tasks, contributing to a 45% burnout rate and a $3 trillion global problem. Nixi AI listens, understands, and completes clinical notes, billing codes, and summaries, saving doctors an hour daily.

Nixi AI employs federated learning to keep patient data on-premise while improving AI models, outperforming standard medical models by 10%. Ms. Yarahmadi emphasized that Nixi AI offers workflow integration and regulatory compliance, unlike general AI tools like ChatGPT, which are not GDPR compliant for healthcare data. The platform provides insurance codes and patient letters, ensuring comprehensive time savings. Nixi AI is live in seven pilot centers in Germany, has completed 3500 sessions, and is raising a €5000 pre-seed round for expansion.

Sean Conaty, CEO and Founder of HealthStay.io, presented an AI enterprise software for health tourism. He explained that hospitals and clinics struggle with manual workflows for international patients, leading to millions in lost revenue and poor patient experiences. HealthStay.io offers a three-layered system: AI agents for qualifying prospects, a vertical CRM for managing the patient journey, and an admin/reporting system for performance visibility.

HealthStay.io achieved a 33% revenue uplift for King’s College Hospital in 12 months, recouping $3 million from a $15 million leakage. The solution automates 90% of the process, centralizes data, and enhances patient engagement by providing immediate answers via AI agents, bringing human interaction at the patient’s request. Mr. Conaty stressed that hospitals need control over patient data and treatment plans, which traditional tourist operators cannot provide due to compliance and legal reasons. HealthStay.io has raised $1.5 million in seed capital and has strategic partnerships in Dubai and Abu Dhabi.

Joe Fernandez, Founder and CEO of Alpaca, shared his personal journey with undiagnosed dyslexia to introduce a platform for early intervention. He noted that dyslexia diagnosis typically takes five years, leading to low self-esteem, parental stress, and higher remediation costs for schools. Alpaca addresses this by having children play evidence-based games three times a year during their first two years of formal education.

Alpaca’s system reduces the time a special education teacher spends on screening from one hour per child to 20 minutes for six children. It combines quantitative game data with qualitative input from teachers, parents, and students to build a comprehensive data spine. Mr. Fernandez stated that Alpaca aims to provide a dyslexia diagnosis in half the time and cost of current methods. Currently, one in four children entering school in Ireland uses Alpaca, leading to earlier interventions and reduced later support needs.

Mr. Fernandez clarified that Alpaca’s customers are school systems and local authorities. While currently focused on dyslexia, the platform will expand to other neurodivergent conditions. Adhering to the EU AI Act’s “human in the loop” principle, Alpaca uses machine learning to infer potential dyslexia but relies on educational psychologists for the final, legally binding diagnosis, validated by input from teachers, special education teams, and parents. This longitudinal tracking ensures evolving support for children.