AI in Healthcare: From Hospital Agents to Healthcare Delivery
Start: 26.11.2026 @ 18:00
Vezilka, together with PyData Skopje and Base42, invites you to an evening dedicated to AI in healthcare. Explore how AI agents can support hospital teams and how operational AI can improve healthcare delivery, with practical examples, live demonstrations, and an honest look at where these systems fail.
When and Where
Date: Thursday, November 26, 2026
Time: 18:00 CET
Location: Base42, Skopje
Sessions
From Chatbot to Care Team: AI Agents in the Hospital
Speaker: Fisnik Doko
Live demos with Microsoft Foundry and Claude, including where they fail
Patients already arrive with answers from AI chatbots, and AI is entering clinical workflows faster than hospitals can evaluate it. The next wave goes beyond answering questions: agents that retrieve guidelines, draft referral letters, and prepare discharge summaries, with a clinician always approving the result.
No technical background needed. Watch ready-made AI assistants work through realistic cases on synthetic data, see where they fail, and leave with a practical roadmap for a safe hospital pilot.
Operational AI in Healthcare Delivery
Speaker: Dimitar Peshevski
Most clinical AI research centres on diagnosis, but the work that consumes healthcare capacity is operational: sorting referrals, producing reports, managing appointments, etc. This talk presents three AI systems built at Sorsix as extensions of Pinga, a configurable platform for healthcare delivery: referral stream processing and extraction, speech recognition and voice-driven editing for radiology reporting, and conversational agents for clinical appointment management. Operational AI has a different evaluation problem from diagnostic AI: the question is rarely whether a model is right in isolation, but whether the surrounding workflow degrades safely when it is wrong.
About the Speakers
Fisnik Doko
Fisnik Doko is a Microsoft MVP in AI (Microsoft Foundry), Cloud & AI Architect, and Assistant Professor at the Faculty of Informatics, Mother Teresa University in Skopje, holding a PhD in Data Science.
With extensive experience in cloud computing, artificial intelligence, and enterprise architecture, he specializes in designing and implementing AI-powered solutions using Microsoft Azure, Microsoft Foundry, and agentic AI technologies. As a Cloud & AI Lead at Vector8, he focuses on building secure, scalable, and responsible AI solutions for enterprise environments, particularly in regulated industries.
Fisnik is also a Microsoft Certified Trainer and an active international speaker, sharing his expertise at Microsoft AI Tour events, technical conferences, and academic institutions across Europe and beyond.
In his session, “From Chatbot to Care Team: AI Agents in the Hospital,” he will explore how AI agents can support healthcare professionals through practical demonstrations using Microsoft Foundry and Claude, highlighting both their capabilities and limitations, with a strong emphasis on safety, governance, and human oversight.
Dimitar Peshevski
Dimitar Peshevski is a Research and Teaching Assistant at the Faculty of Computer Science and Engineering at Ss. Cyril and Methodius University in Skopje and a Research Scientist at Sorsix. His work connects academic research in artificial intelligence with practical applications in healthcare delivery.
He earned both his BSc in Computer Science and MSc in Data Science with a perfect GPA of 10.00 and was awarded the Gold Engineering Ring for outstanding academic achievement. He is currently pursuing a PhD in Computer Science and Engineering.
At Sorsix, he researches and develops AI solutions to support healthcare professionals and improve the delivery of care. His work focuses on translating advances in AI into practical tools that streamline operations, reduce administrative workload, and integrate reliably into healthcare workflows.
Registration
Organizers: Vezilka, PyData Skopje and Base42.
Register:
There is limited seating, so please register for the event if you would like to attend.
Base42 was made from scratch by enthusiasts like you.
© 2042