Fwdays+DevRain AI 2025
2025-07-05
-- online --
who it's for
machine-learning engineers, ai engineers, product managers, software engineers, research engineers, devops engineers, data engineers
topics
LLM agents validation; observability for agents; computer-use agent training; SFT to PPO recipes; agent-based mobile apps; video generation pipelines; multilingual RAG systems; Ukrainian embeddings evaluation; hallucination scoring metrics; human-AI collaboration models; agent autonomy limits; investment strategies for AI infrastructure
| title | author | length |
|---|---|---|
| Validation and Observability of AI Agents | Oleksandr Denysiuk | 47m |
A prototype mobile app called MyA connects to a multiagent backend using OpenAPI and Crew AI with FastAPI, a vector database for long-term memory, and… | ||
| Computer-Use Agents: From SFT to Classic RL | Maksym Shamrai | 42m |
Describes concrete computer-use agent systems and datasets, including OpenAI and Anthropic Claude computer-use APIs, Code Agent beta, and the Android Wild… | ||
| Generative AI: A Google Perspective | Tetiana Lukyniuk | 29m |
Overview of Google’s perspective on generative artificial intelligence, covering recent developments and practical applications from Google’s research and… | ||
| Autonomy of LLM Agents: Current State and Prospects | Oles Petriv | 50m |
Examines what currently works for autonomous LLM agents, the technical and practical limitations and their root causes, and research directions for extending… | ||
| Video Generation + AI: From Production to Product | Vasil Goshovskiy | 39m |
The talk explains a practical AI video-production pipeline using Midjourney/Fripik for detailed stills, LoRA fine-tuning to refine characters, VO3/Runway for… | ||
| Should You Invest in AI in 2025? | Oleksandr Sharko | 38m |
Overview of the current state of investments in AI in 2025, covering which AI products and sectors are attracting funding and whether it makes sense to invest… | ||
| Beyond English: Challenges of Building a Ukrainian-Language RAG System | Roman Romanov | 34m |
Examines challenges of building a Retrieval-Augmented Generation (RAG) system for Ukrainian, focusing on evaluation methods in domain-specific settings such as… | ||
| What I Learned Writing a Book About Generative AI Using Generative AI | O. Krakovetskiy | 45m |
A developer describes writing a book about generative AI with the help of generative-AI tools, covering ethical dilemmas, cognitive challenges, and practical… | ||
| Human-AI Collaboration Models for Better Decisions and Faster Workflows | Oleksii Minakov | 40m |
Presents concrete human–AI interaction models (centaur, cyborg, ensemble, author-ghost, resonator, agent) and gives criteria for selecting among them based on… | ||
| The Wow-Case Competition | Oleksandr Krakovetskiy, Oleksii Minakov | 45m |
Alexey Minakov and Oleksandr Krakoveckyi compete in a live showcase of striking and unexpected use cases for generative AI tools, demonstrating creative… | ||
| AI Is Already Here. What Will Happen to Your Team (and Your Role) Tomorrow? | Artem Bykovets | 50m |
Generative AI is reshaping development teams by delegating many tasks to assistants, multi-agent systems and copilots, leading to smaller teams but higher… | ||
| nothing matches | ||