The 11th international conference on Machine Learning and Artificial Intelligence applications,
taking place in person in Prague and online.

Machine Learning Prague 2026

– , 2026

Registration

World class expertise and practical content packed in 3 days!

You can look forward to an excellent lineup of 45 international experts in ML and AI business and academic applications at ML Prague 2026. They will present advanced practical talks, hands-on workshops, and other forms of interactive content to you.

What to expect

  • 1000+ Attendees
  • 3 Days
  • 45 Speakers
  • 10 Workshops

Phenomenal Confirmed speakers

Weipeng Xu

Senior Research Scientist, Valka.AI, ex Meta

Weipeng Xu, Ph.D. is a research scientist specializing in human-centric computer vision, graphics, and machine learning. He has led pioneering work in digital humans, neural rendering, and motion capture, with extensive experience spanning academia and industry, publishing widely in top venues such as CVPR, SIGGRAPH, and ICLR. He is currently a Senior Research Scientist at Valka.AI, leading research on digital human video generation using diffusion models. Previously, he was a Staff Research Scientist at Meta, driving innovations in egocentric human pose estimation, Codec Avatar telepresence, and generative human motion modeling. Before joining Meta, he was a Postdoctoral Researcher at the Max Planck Institute for Informatics, where he advanced human motion capture, 3D reconstruction, and neural rendering. His work has consistently bridged fundamental research with real-world applications in digital humans, computer vision, and machine learning.

Ariane Mora

Researcher, Caltech

Tomas Pluskal

Research Group Leader, Institute of Organic Chemistry and Biochemistry of the Czech Academy of Sciences

Tomáš Pluskal is a group leader at IOCB Prague, where he heads the Biochemistry of Plant Specialized Metabolites research group (www.pluskal-lab.org). Building on his Ph.D. training from the Okinawa Institute of Science and Technology and a postdoctoral fellowship conducted at the Whitehead Institute for Biomedical Research, he develops computational and biochemical approaches to discover bioactive plant molecules. He co-authored the widely used software platform mzmine and recently introduced the AI-driven DreaMS model for identifying unknown compounds from mass spectrometry data. Tomáš is a recipient of the Neuron Award for Promising Scientists, an ERC Consolidator Grant, and is a member of the EMBO Young Investigator Programme.

Alexandr Vendl

Team Lead and AI Architect, O2/Dataclair

As the team lead and AI architect at Dataclair/O2, Alexandr Vendl is in charge of the AI Agentic systems used in their contact center. He leads the team that puts these complex AI systems into production, where they manage thousands of user interactions every day with an emphasis on monitoring and evaluation. Since joining the company in 2022, his work has evolved from focusing on the early adoption of Large Language Models and fraud detection to architecting these larger, more sophisticated solutions. His approach is shaped by a strong background in mathematical modeling and deep learning, stemming from his PhD research into ultrafast laser systems.

Practical & Inspiring Program

Friday
Workshops

O2 Universum, Českomoravská 2345/17a, 190 00, Praha (workshops won't be streamed)

Registration from 8:00

Room D2 Room D3 Room D4 Room D6 Room D7
– coffee break
–

Graph-Based Anomaly Detection: Methods and Hands-On Implementation

Room D2

Adam Jurčík, Seznam.cz
Jakub Chynoradský, Seznam.cz
Jaroslav Kuchař, Seznam.cz

This workshop introduces participants to the fundamentals and practical techniques of graph anomaly detection. Blending conceptual overviews with a hands-on lab the session teaches how to represent real-world systems as graphs engineer meaningful features and apply a range of detection methods to identify unusual nodes edges or communities. We will provide a concise overview of modern GAD methods categorized into the following families: Traditional Methods Embedding-Based Methods and Graph Neural Network (GNN).

Train a Small Language Model with SageMaker HyperPod

Room D3

Adir Gozlan, Amazon
Mayank Gupta, Amazon

In this highly practical workshop participants will experience end-to-end training of a small-scale language model using Amazon SageMaker HyperPod a cutting-edge distributed training environment purpose-built for large foundation models. With growing interest in domain-specific and lightweight LLMs this session will bridge the gap between theoretical knowledge and applied training workflows on modern cloud infrastructure. Participants will leave with a reproducible cloud-native setup and hands-on familiarity with distributed LLM training — from resource allocation to performance tuning. We will emphasize how to make small models perform competitively using techniques like LoRA quantization and mixed-precision training.

Probabilities Over Prompts: How to Stop Begging and Start Modeling

Room D4

Ludek Matyska, Rouvy
Vojtech Kadlec, STRV

Modern interaction with large language models often ends with a plea: “please don’t hallucinate.” This workshop revisits the foundations of machine learning and shows how they remain essential in the era of generative AI. Participants will learn to work with the probabilistic structure that LLMs expose using token-level log probabilities to build classifiers evaluate uncertainty and detect hallucinations. Through practical examples we demonstrate how classical ML techniques such as calibration fine-tuning and systematic testing can produce reliable verifiable outcomes. Rather than treating LLMs as black boxes to be prompted we treat them as probabilistic learners to be measured — because when a model hallucinates the answer is not to beg it but to calibrate and test it.

From Prediction to Impact: Machine Learning Meets Econometrics for Causal Inference in Organizations

Room D6

Nadzeya Laurentsyeva, Sanofi
Luděk Stehlík, Sanofi
Ivan Ushakov, Sanofi

Organizations increasingly rely on data-driven insights to evaluate the impact of various initiatives and programs yet randomized controlled trials (RCTs or A/B tests) often prove infeasible due to logistical ethical or political constraints. The workshop employs a hands-on case study approach to guide participants through the complete impact evaluation pipeline. Using a realistic case from people analytics (impact evaluation of an employee training program) participants will navigate four critical stages: research design selection of appropriate identification strategies implementation of advanced estimation methods and effective communication of results to stakeholders. The case study presents two distinct scenarios: a staggered rollout with random wave assignment and a voluntary global rollout—illustrating how design constraints shape methodological choices and the rigor of causal claims.

From Maps to Models: Hands-On Geospatial Data Processing for Data Scientists

Room D7

Filip Kojan, O2/Dataclair
Ondřej Karlíček, O2/Dataclair

This workshop offers a comprehensive introduction to geospatial data processing tailored for beginner and intermediate machine learning practitioners. Participants will explore essential concepts including spatial data types coordinate reference systems and projections as well as core operations for analyzing and manipulating geospatial information. You will see how geospatial data can be integrated into machine learning workflows through feature engineering and spatial analysis. Additionally visualization techniques will be covered to help interpret and communicate geospatial data.

–
Lunch
– coffee break
–

Find the Bottleneck: Speed Up ML Pipelines by 10%–500%

Room D2

Michal Šustr, FEE CTU

Modern GPUs keep getting faster each generation enabling ever larger training and inference workloads. But in many real-world applications GPU compute is no longer the slowest part: I/O and data movement are. This workshop teaches a practical systematic approach to diagnosing and eliminating data-loading bottlenecks. Participants will learn how to design high-throughput data pipelines for both training and inference. By the end of the session attendees will be equipped with techniques that typically yield 10% to 500% pipeline speedups — often without touching the model architecture at all.

From Zero to Task-Master: Hands-On Fine-Tuning of Large Language Models for Specialized Applications

Room D3

Elad Ben-Zaken, Dream
Oded Ovadia, Dream

Large Language Models (LLMs) such as GPT Claude and Grok have transformed the landscape of AI but using them effectively for specific real-world tasks often requires more than prompt engineering. Fine-tuning task-specific adaptation and efficient training strategies are essential to unlock their full potential. This workshop is designed for advanced practitioners who want to go beyond using LLMs as generic assistants and learn how to systematically adapt them to rule a particular domain or task. Over the course of 3 hours participants will gain practical hands-on experience in training and adapting LLMs. Starting with a pre-trained foundation model we will walk through the full pipeline of dataset preparation fine-tuning evaluation and deployment. Attendees will bring their laptops and work along implementing techniques directly in code.

Building and Backtesting ML Investment Strategies

Room D4

Karolina Cynk, CERN

This workshop provides a technical walkthrough for constructing and evaluating machine learning – based investment strategies using fundamental financial data (information about a company’s performance e.g. revenue profits and debt). Drawing on my practical experience at the CERN Pension Fund we will cover the full development pipeline: data preprocessing signal generation and backtesting.

Beyond Single Decision: A Practical Tutorial on Slate Bandits

Room D6

Roman Dušek, Bloomreach
Nairit Banerjee, Bloomreach

This hands-on workshop is designed for machine learning practitioners and data scientists seeking to advance from single-item personalization to solving complex multi-slot decisions. While contextual bandits excel at selecting one optimal item modern digital platforms require the simultaneous optimization of entire "slates" of content. This session provides a comprehensive overview of the theory and practical application of slate bandit algorithms.

Recommender Systems in the Age of Generative AI

Room D7

Jan Kislinger, Sky Czech Republic

Generative AI has revolutionized content creation but what if we use it to generate the entire recommendation experience itself? In this workshop we explore how generative models can go beyond tagging or summarizing content to compose full recommendation pages. Participants will build prototypes that generate coherent page layouts using autoregressive models that attend to previously generated tiles. We will connect traditional recommenders with new generative methods to show how to move from ranking individual items to creating personalized experiences end to end.

Saturday,
Workshops

O2 Universum, Českomoravská 2345/17a, 190 00, Praha (and on-line)

Registration from 9:00

Program will be announced.

Sunday,
Conference day 1

O2 Universum, Českomoravská 2345/17a, 190 00, Praha (and on-line)

Doors open at 08:30

Program will be announced.

Have a great time Prague, the city that never sleeps

You can feel centuries of history at every corner in this unique capital. We'll invite you to get a taste of our best pivo (that’s beer in Czech) and then bring you back to the present day at our networking event.

Venue ML Prague 2026 will run hybrid, in person and online!

The main conference as well as the workshops will be held at O2 Universum.

We will also livestream the talks for all those participants who prefer to attend the conference online. Our platform will allow interaction with speakers and other participants too. Workshops require intensive interaction and won't be streamed.

Conference building

O2 Universum
Českomoravská 2345/17a, 190 00, Praha 9

Workshops

O2 Universum
Českomoravská 2345/17a, 190 00, Praha 9

Now or never Registration

Early Bird

Only 50 Registrations Left

Standard

600 Registrations Until March 31

  • Conference days € 320
  • Only workshops € 260
  • Conference + workshops € 540

Late

Until sold out

  • Conference days € 350
  • Only workshops € 290
  • Conference + workshops € 590

What You Get

  • Practical and advanced level talks led by top experts.
  • Networking and drinks with speakers and people from all around the world.
  • Delicious food and snacks throughout the conference.

They’re among us We are in The ML Revolution age

Machines can learn. Incredibly fast. Faster than you. They are getting smarter and smarter every single day, changing the world we’re living in, our business and our life. The artificial intelligence revolution is here. Come, learn and make this threat your biggest advantage.

Our Attendees What they say about ML Prague

Thank you to Our Partners

Strategic Partners

Platinum Partners

Communities and Further support

Would you like to present your brand to 1000+ Machine Learning enthusiasts? Send us an email at info@mlprague.com to find out how to become a ML Prague 2026 partner. Our basic partnership offer can be found here.

Become a partner

Happy to help Contact

If you have any questions about Machine Learning Prague, please e-mail us at
info@mlprague.com

Organizers

Jiří Materna
Scientific program & Co-Founder
jiri@mlprague.com

Teresa Pulda
Event production
teresa@mlprague.com

Gonzalo V. Fernández
Marketing and social media
gonzalo@mlprague.com

Jona Azizaj
Partnerships
jona@mlprague.com

Ivana Javná
Speaker support
ivana@mlprague.com

Barbora Toman Hanousková
Communication
barbora@mlprague.com

Jan Romportl
Moderator