April 10, 2026
The future is here. Explore artificial intelligence, large language models, computer vision, AI agents, and cutting-edge ML techniques. Learn how Python powers the AI revolution and how you can be part of it.
Keynote Speakers
Piotr Skalski
Open Source Lead, Roboflow
Piotr is Open Source Lead at Roboflow, building tools for computer vision developers. With 8 years in computer vision, he's created open source projects with over 70k GitHub stars, written 40+ blog posts, and produced 60+ YouTube videos covering key computer vision models. Known for his sports-focused projects including Football AI and Basketball AI.
Geoffrey Huntley
Open Source Engineer, Independent
Geoffrey Huntley is the creator of the Ralph Wiggum Loop, a brute-force AI agent technique that turns simple prompts into fully functional software through automated iteration. His recent work includes Loom (an AI orchestrator/software factory as an evolution of Ralph), the 'Cursed' programming language (a compiler written entirely by AI loops), and the 'How to Build a Coding Agent' workshop. Known for his viral activism with The NFT Bay, Geoff brings a unique mix of deep engineering rigor and chaotic creativity to the stage. He currently lives remotely on a goat farm in Australia.
What to Expect
Talks and workshops happening on AI Day
AI Day Intro
AI Lithuania Summit
Linas Petkevičius
Head @Instutute of computer science, Vilnius university President of Artificial intelligence association of Lithuania
Reading the Mind of an LLM
Luca Baggi, Gabriele Orlandi
AI engineer @xtream and open source contributor, AI scientist at xtream
Reading the Mind of an LLM
What if you could watch an AI’s thought take shape? For years, LLMs have been impenetrable "black boxes," but we are finally beginning to find ways to see how the ghost in the machine actually works. This talk explores **mechanistic interpretability**, a subfield of AI that aims to understand the internal workings of neural networks. Mapping these internal "circuits" is not only just a philosophical curiosity - or duty: it is a high-stakes engineering necessity for safety, debugging, and trust.
Luca Baggi, Gabriele Orlandi
AI engineer @xtream and open source contributor, AI scientist at xtream
Measuring Experiments in LLMs: A/B Tests and Automated Testing
Even small changes in LLMs can impact output quality, safety, and user experience. In this talk, we’ll show how to log experiments with Langfuse, automate tests with Pytest, and enrich them using Hypothesis-generated random data scenarios. Participants will learn how to use code, tests, and data-driven A/B tests to improve LLM development.
Özge Çinko, Kader Miyanyedi
Hello World, I'm Özge Çinko! 👋 I'm a computer engineer who finds inspiration at the intersection of curiosity and technology. Currently building the future as an AI Engineer at ING. For me, engineering is a creative craft - turning data into narratives and emotions into visual experiences. I am passionate about making technology more human-centric and purposeful. When I'm not coding, I'm usually writing, traveling, or chasing the thrill of learning something new., I have been a backend developer for 4 years, working primarily with Python and Django. I enjoy sharing what I’ve learned at previous PyCon talks and through writing on Medium, helping others improve their coding and AI skills.
Context Engineering with DeepAgents: Write, Select, Compress, Isolate
Context Engineering is the new prompt engineering. As agentic workflows grow, just appending messages to a list causes context poisoning and latency spikes. In this talk, we'll look at a better architecture using the `deepagents` library. We'll explore the four pillars of context engineering: Write, Select, Compress, and Isolate. You'll learn how to treat context as a finite resource and build autonomous agents that can solve complex tasks without crashing your context window.
Antanas Daujotis
With over 20 years in the trenches of engineering, he now focuses on the messy reality of getting agentic systems to work in production.
AI-Assisted Development in Practice: Chatbot + Agentic Testing from Scratch
GenAI-based projects are easy to prototype, but hard to test. In this hands-on workshop, we will build a simple AI chatbot from scratch and, in parallel, create an AI agent that tests it during development. The testing agent will generate user scenarios, detect failures such as prompt issues, incorrect tool usage, etc., and then suggest concrete next steps for improvement. The core theme of the workshop is building AI systems with the help of AI itself, while keeping an engineering-driven approach to design
Alex Marmuzevich
I am a software professional and solution architect with 30+ years of experience in IT. I have been developing in Python since 2010. Prior to that, I worked extensively with ASM, C/C++, C#, etc. I have been actively using AI in software development since 2023. At present, up to 80% of my production code is written with the help of AI agents. My main focus is on applying AI-assisted development practices in a disciplined, engineering-driven way. I am an AI Ambassador at EPAM, where I promote practical adoption of AI tools and workflows in everyday software engineering.
AI Lithuania Summit
Tutorial: Unsloth for Small Language Models Fine-Tuning Small Language Models (SLMs) can deliver strong performance with far lower computational demands than large LLMs, making them ideal for on-device, edge, or cost-sensitive applications. However, fine-tuning them effectively and quickly on limited hardware remains challenging. This hands-on tutorial on Unsloth, an open-source library that makes fine-tuning SLMs dramatically faster and more memory-efficient.
Linas Petkevičius
Head @Instutute of computer science, Vilnius university President of Artificial intelligence association of Lithuania
What It Means to Be a CTO in an AI Startup Today
**I haven't written a single line of code in over a year.** As CTO of an AI startup, I've fully embraced vibe coding: orchestrating AI to generate production-ready code. 🔍 **Highlights** 1. **The paradigm shift:** from autocompletion to full project generation 2. **My 3-step workflow:** voice prompts, deep research, code generation 3. **The judgment trap:** why expertise matters more than ever You'll leave with practical techniques to ship 10x faster!
Fabien Vauchelles
Fabien Vauchelles brings 20+ years of experience in product design and software development. He holds a dual degree from Université Paris-Sud in Engineering and Master Research in Distributed Systems. A serial entrepreneur, Fabien has founded three companies including Zelros, where he served as CTO for six years building the AI platform from scratch. He is now CTO and Co-founder of an AI startup, applying vibe coding principles daily. Recognized internationally as an expert in AI, distributed systems, and data science, Fabien has spoken at conferences in 15+ countries on topics including web scraping, anti-bot systems, and open-source tooling. He is also the creator of Scrapoxy, a popular open-source web scraping tool.
AI & Ethics
It's been in the news, everywhere. Artificial Intelligence is the next thing in the evolution of technology. While it's probably not the end of everything as some sources say, it's likely true to change the face things for years to come. In this talk, we'll talk about what developers should focus on to build their AI knowledge and skills, what things we need to know outside of the technology to help our understanding, and some general ideas around AI itself.
PJ Hagerty
PJ Hagerty is a well-known figure in the tech industry, particularly within the developer relations and DevOps communities. He is recognized for his work as a developer advocate, community builder, writer, and speaker. PJ founded DevRelate.io, a company focused on helping tech organizations build communities and enhance developer relations. He has a strong presence in the open-source and tech communities, often speaking at conferences and producing content on topics such as software development, DevOps, and community engagement.
Attacking Toughest and Messiest Tech Challenges with AI
Legacy migrations, tangled integrations, undocumented business logic – every team has that category of work nobody wants to touch. This talk shares practical lessons from using AI to tackle exactly those challenges. Through the example of a real IT system migration – normally a 2–4 year project – we will show how AI can compress the timeline to weeks, reduce the pain, and even make it fun.
Egidijus Pilypas
Egidijus Pilypas is Co-founder and Product Director at Exacaster. He has been working with AI and machine learning for over 15 years, spanning telecom, retail, finance, and utilities - long before it became a buzzword. His focus has always been the hard end of the work: the messy data, the legacy systems, the delivery challenges that others avoid. At PyCon Lithuania 2026, Egidijus will share hands-on lessons from using AI to tackle exactly those kinds of challenges, and why the toughest projects might now be the most interesting ones.
Vibe reverse engineering of old games and new hardware
Reverse engineering binaries once required deep expertise. Today, AI models like Opus 4.6, GPT-5.3-Codex, and Gemini 3.1 Pro change the rules. Watch how pairing AI with the NSA's Ghidra decompiler or simple hex tools like xxd makes binary hacking accessible. We will dive into practical projects: hacking infinite lives into Atari’s River Ride, porting the legacy game Chromatron, decoding LED backpack protocols, and hunting for backdoors. Let me show how to add reverse engineering to your everyday skills.
Piotr Migdał
Piotr Migdał /pjɔtr ˈmig.daw/ - a curious being, doctor of sorcery. Professionally: I am a founding engineer at Quesma, investigating ever-changing limits of agentic AI in software engineering. Previously: Co-founder & CTO of Quantum Flytrap, deep learning consultant, data viz specialist, quantum physics PhD. Personally: I dance balfolk, fusion, and Zouk. I do sauna rituals, among trees, ponds and streams. A bit more on my journey in this post.
XAI - Explainable AI tools and techniques
In this talk, I propose to discuss the problem of building explainable AI with the two approaches - causal vs correlational. I will talk about what mech interp in LLMs. As a way to understand how models answer questions by looking inside them and checking which neurons activate when. I will discuss Anthropic's open sourced a python module - circuit-tracer, the Neuronpedia portal , will also talk about my own work with "activation cube" data structure (this is not a standard - I came up with it)
Viraj Sharma
I am a passionate technologist with a strong interest in Python, artificial intelligence and Edge computing. I am currently studying in Class 9 at Presidium School, Delhi, INDIA. I have worked in areas such as torch.nn visualization, Anthropic technologies (MCP, Skills), large concept models, and TensorCore/CUDA benchmarks, Edge AI on raspbrry pi running small models with sensors. Recently I have been working on XAI (AI explaianability) and putting my work as a project on my AI Lab - modelrecon.com. As an active member of the Python community and AI communities, I enjoy learning from experienced developers and sharing my insights with others. I attend major tech events including PyCons, Linux Fests, OS Summits, GDG events, p99conf, and various AI conferences, where I actively present my projects and ideas.
Beyond Basic RAG: Boosting Accuracy with Hybrid Search and Fusion Algorithms.
Retrieval-Augmented Generation (RAG) often produces incorrect results when relying solely on vector-based similarity. While vector search is strong, it fails on exact keywords, acronyms, and domain-specific terms. This talk shows how to build high-accuracy RAG pipelines with Hybrid Search and Re-ranking using LangChain, combining vector and full-text retrieval and applying cross-encoder scoring to deliver more relevant context and reduce hallucinations.
Piti Champeethong
I've been working with databases and software development for 20 years. Currently, I'm a MongoDB senior consulting engineer based in Singapore. I've previously spoken at conferences such as PyCon Lithuania 2025, PyCon APAC 2025, PyCon SG 2025, PyCon Thailand 2025, and Global Azure Thailand 2025. I’m also part of the community leader team for the MongoDB and PyLanna (the Python) User Group in Thailand, which brings together over 3,000 developers.
AI Agents: risks and legal
With AI Agents there is multiple things from risks we need to take into account to, legal issues which might appear. We will have a panel discussion with Moderator: prof. Paulius Pakutinskas Panel Jonas Lekavičius, Du Bitai Karolina Griciūnė, SwirlAI Neringa Gaubienė, VU
Linas Petkevičius, Karolina Griciunė, Jonas Lekevicius, prof. Paulius Pakutinskas, Neringa Gaubiene
Head @Instutute of computer science, Vilnius university President of Artificial intelligence association of Lithuania, TBD, Host of "Du Bitai" podcast / Co-founder of Lithuanian Artificial Intelligence Association / Vibecoder at Silent, Prof. Paulius Pakutinskas is a law professor and AI governance expert based in Lithuania, working at the intersection of regulation, innovation, and societal impact. He leads the Legal Tech Centre at Mykolas Romeris University and holds a UNESCO Chair on Artificial Intelligence, Emerging Technologies and Innovations for Society. He is a member of the European Working Group of Competent Authorities on AI and heads a subgroup focused on general-purpose AI, contributing to the implementation of the EU AI Act. His work bridges academia, policy, and practice, with a strong focus on how AI can be deployed responsibly, securely, and at scale — particularly in the public sector and justice systems.,
Beyond the Prompt: Building Research Agents in Python
LLMs can answer questions, but can they conduct research? Real research requires planning, data gathering, synthesis, and refinement. This talk explores Python patterns for orchestrating multi-step workflow to create autonomous research agents. We'll deconstruct the agentic architecture, from DAG-based planning to reflection loops, and show how to implement these concepts using practical Python patterns.
Simona Skiotytė
As a Software Developer for the Generative AI platform at Nasdaq, Simona Skiotytė specializes in architecting the integration of Generative AI systems on AWS. Key projects include the integration of a research system that automates query generation, iterative web search, and multi-stage report synthesis, all orchestrated in AWS. Simona's foundation in data science comes from her time at ETH Zürich, where she applied statistical and mathematical models to large-scale biological datasets, utilizing dimensionality reduction, clustering, and regression modeling. This background provides a strong analytical lens through which Simona approaches the construction of Generative AI systems.
Death by a Thousand Prompts: Can Our Disclosure Standards Survive AI Slop?
AI-generated "slop" is overwhelming vulnerability triage and burning out maintainers. This session focuses on building a unified framework to identify and "black-hole" synthetic noise at scale. We will discuss practical, cross-platform strategies to automate the rejection of low-signal reports and protect engineers from the unsustainable volume of AI-augmented disclosures.
Jarek Potiuk
Independent Open-Source Contributor and Advisor, Committer and PMC member of Apache Airflow, Member of the Apache Software Foundation, Security Committee Member of the Apache Software Foundation. Organizer of community-focused events, speaker. Jarek is an Engineer with a broad experience in many subjects - Open-Source, Cloud, Mobile, Robotics, AI, Backend, Developer Experience, Security, but he also had a lot of non-engineering experience - building a Software House from scratch, being CTO, organizing big, international community events, technical sales support, pr and marketing advisory but also looking at legal aspects of security, licensing, branding and building open-source communities are all under his belt. With the experience in very small and very big companies and everything in-between, Jarek found his place in the Open Source world, where his internal individual-contributor drive can be used to the uttermost of the potential.
Stop Guessing: Build Feedback Loop for Prompt Engineering
Most teams iterate on prompts by eyeballing outputs and hoping for the best. This talk presents a different approach: connect your prompt to a measurable downstream outcome, build a test set, and let metrics drive improvements. I'll walk through a complete feedback loop - from prompt to retrieval metrics to automated error analysis to LLM-powered iteration - and show how this turns prompt engineering from guesswork into something you can actually measure and improve systematically.
Tadas Goberis
Dentist turned engineer with 4 years in software and nearly 3 years in AI. Currently Lead AI Engineer at Trimble, where I build production agentic systems.
Quantum Machine Learning with Qiskit
Quantum Machine Learning (QML) combines quantum computing and classical machine learning, but its practical value is often misunderstood. In this hands-on workshop, we will explore QML using Qiskit, build and run real quantum machine learning models, and compare them with classical approaches. The session focuses on practical intuition, runnable Python code, and a clear discussion of current advantages, limitations, and realistic use cases of QML.
Artem Konotpchyk, Manta Ribkauskyte
I am a Data Scientist and AI Engineer at IBM, where I work on machine learning systems and applied AI in production environments. My day-to-day work focuses on building, deploying, and evaluating ML models, while also exploring emerging computational approaches that challenge classical assumptions. Alongside classical machine learning, I work with quantum computing and Quantum Machine Learning, with an emphasis on practical experimentation rather than theory alone. I have delivered a hands-on quantum computing workshop at IBM, introducing quantum concepts through runnable code and real examples. Using Qiskit, I design and test hybrid quantum–classical workflows, compare them with classical baselines, and analyze the effects of noise, limited qubit counts, and current hardware constraints.,
Engineering Complex AI solutions: Observability and Testing of multi-Agent Solutions
As AI agents evolve from simple chatbots to complex multi-agent systems utilizing Model Context Protocol (MCP), manual validation becomes impossible. During this talk, I will demonstrate the process of architecting a quality assurance loop for these solutions. No theoretical fluff: I will focus on the practical analysis of automated pipeline results, interpreting Langfuse reports for cost and performance, and ensuring reliability from an AI Architect/System Engineer perspective.
Dmitri
Lead systems engineer with more than 20 years experience in IT: - 3 years in AI-based solutions development; - 8 years of S-SDLC methodology implementation; - 7 years of security architecture development; - 4 years Experience in the development of the scalabale solution for the non-functional testing using private clouds AWS based APIs - solid knowledge of building secure and resilient architectures in multi-cloud environments - 9 years of experience in consulting services for the external and internal EPAM accounts - 6 years of experience in project and team management - 8 years of experience in solutions architectures development and deployment - 10 years security concepts and tests development including regular security audits - 8 years of experience of developing and deploying Continuous Delivery and Continuous Integration concepts - Experience in processes development - 10 years developing and deploying Unix/Linux based infrastructures
From OpenAI to DeepSeek: New Scaling Laws for LLMs that can Reason
With o1, OpenAI ushered a new era: LLMs with reasoning capabilities. This new breed of models broadened the concept of scaling laws, shifting focus from train-time to inference-time compute. But how do these models work? What do we think their architectures look like, and what data do we use to train them? And finally - and perhaps more importantly: how expensive can they get, and what can we use them for?
Luca Baggi
AI engineer @xtream and open source contributor
From Data Collection to Partner Delivery: Shipping a Paraphrasing Skill on Small Language Models
I'll discuss building capabilities on top of on-device language models, with Rewrite as a case study – a publicly available paraphrasing skill that is widely used across Microsoft's products. I'll cover comprehensive data collection strategies, carefully designed adapters training and evaluation. I'll discuss the engineering challenges we faced: achieving target latency while maintaining quality, hardening the system against edge cases, and how to deliver the technology to partners and the lessons learnt.
Marat Saidov
Marat Saidov is a Senior Software Engineer at Applied Sciences Group, Microsoft. Based in Belgrade, Serbia. Previously improved Speech Recognition and Natural Language Understanding services at Alice Voice Assistant, Yandex. Besides that, he was an NLP Research Assistant at HSE University, Russia.
AI and Agency: As Developers, We Decide The Future
AI systems are not neutral; they encode values that are opaque and separate from people. As developers, we are making decisions that shape who has power, who is surveilled, who is replaced, and who gets a voice, often without intending to. This talk frames AI ethics as a software engineering problem, not just a philosophical one. We’ll examine how everyday technical choices can unintentionally reinforce authoritarian tendencies around disinformation, manipulation, and harmful automated decision making.
Tadas Korris
Tadas Korris is a software engineer at Mozilla, currently working on the Sync Backend Storage team, which is responsible for safely and securely synchronizing browser data for millions of Firefox users. His work focuses on building reliable, privacy-preserving systems that users can trust with their most sensitive data. Previously, Tadas worked on Mozilla’s Contextual Services team, where he contributed to the Merino service, which provides private, contextual suggestions while maintaining strong privacy guarantees. Across his roles, he has been a strong advocate for online privacy and security, with hands-on experience developing secure software throughout its lifecycle. Tadas presented at PyConLT in 2024, discussing the Merino service in detail. Tadas is deeply engaged in the ethics of AI and emerging technologies and has given talks on the dangers of unchecked automation and the steps technologists can take to protect democratic institutions and social trust in tech. He has presented this work at Mozilla Festival 2025 in Barcelona, the IIA International Conference in Amsterdam, and various conferences in Canada. Tadas was born to a Lithuanian-Canadian family in Toronto and grew up in Edmonton. He maintained close ties to the local Lithuanian community, learning the language and participating in cultural and community activities from an early age. He cherishes his Lithuanian heritage and has gotten quite good at making vegan versions of almost all Lithuanian dishes. His Močiute Emilija would be proud! Before transitioning into software engineering, Tadas began his professional career as a classical musician. He earned both his Bachelor’s and Master’s degrees from the Manhattan School of Music in New York City. In 2018, he completed a diploma in Web and Software Development from the University of North Carolina and began working in the technology sector. He joined Mozilla in 2022 and continues to perform as a regular substitute musician in several orchestras.
Friend or Foe? AI at Play in Cybersecurity
AI is a double-edged sword in cybersecurity. This talk explores its dual role. Why AI excels: Vast open-source training data and profit-driven, coding-optimized models make AI a fast, multi-domain expert at writing and finding vulnerabilities in code. Gatekeeping: Projects like Claude Mythos and Project Glasswing raise hard questions about who should access these powerful tools. The asymmetry: AI is fundamentally reshaping the defense/offense balance—and demands responsible deployment.
Cheuk Ting Ho
After having a career as a Data Scientist and Developer Advocate, Cheuk dedicated her work to the open-source community. Currently, she is working as a developer advocate for JetBrains. She has co-founded Humble Data, a beginner Python workshop that has been happening around the world. Cheuk also started and hosted a Python podcast, PyPodCats, which highlights the achievements of underrepresented members in the community. She has served the EuroPython Society board for two years and is now a fellow and director of the Python Software Foundation.
LLMs through my last 3 professions: an interdisciplinary approach
We have heard a lot about LLMs from tech experts, but what do an economist, an English teacher, and a tourism manager have to say about it? Fortunately, I've been all three and would love to share my experience!
James Donahue
Raised in Nashville, Tennessee, USA, I have bounced around the world, finally settling in Hamburg, Germany. My professional background reflects my vagabond nature. I taught English in Asia, Latin America, and online (full-time remote work before Covid!), worked in adventure tourism in Appalachia (USA) and Chile, and did my masters in Economics in Hamburg. Somewhere along the way, I gathered a few stories and soft skills. I met Python during my masters, but stuck with Matlab through the my PhD coursework, until I decided that academia is not my forever home. Currently I am occupying myself with ensemble methods such as XGBoost, as well as expanding into computer vision with PyTorch and toying with more advanced data visualizations. Naturally, NumPy will always hold a special place in my heart, along with the Cython package to sample everything a Bayesian needs.
