Why This Session Matters
In this session, senior engineers from deepsense.ai unpack what actually works when you need agents to perform under real traffic and real constraints. In just 60 minutes, you’ll learn how to:
- Move from clever demos to durable, observable agent systems
- Cut manual resolution time by 40–70% with structured orchestration and memory
- Reduce inference costs by up to 35% via caching, routing, and guardrails
- Design evaluation, telemetry, and scorecards for multi-agent reliability
- Avoid common failure modes (tool use, context bloat, retries) and stabilize latency
This program is regularly offered to global clients as a paid workshop, and we are sharing a condensed, 60-minute version in this session.
What’s Inside
In this workshop, in 60 minutes, we covered:
Architectures & Frameworks
(LangGraph, ragbits-style patterns, deterministic sub-agents)
Core Capabilities
(evaluation pipelines, shared memory, observability, regression tests)
Reliability & Cost Controls
(routing, retrieval caching, temperature/decoding policies)
Case Studies
(developer-assist agent, multilingual voicebot, and pharma document intelligence)
No sales pitches — just tested practices and frameworks.
What You’ll Receive
- Instant access to the recording (on-demand)
- Executive Summary with key takeaways and patterns to reuse
- Reference architectures & checklists used in enterprise rollouts
Meet the Speakers
Rafał Łabędzki
AI Engineering Manager, deepsense.ai
As an AI Engineering Manager at deepsense.ai, Rafał leads the development of intelligent systems across various industries, including in-silico drug discovery, FMCG, manufacturing, and telecommunications, for global enterprises.
His career began in the video game industry in 2009, where he built a foundation in complex systems and user-centric design. Today, he combines deep technical expertise with academic research: Rafał holds master’s degrees in finance and management, as well as a PhD in management. His research focuses on human-AI collaboration in hybrid teams.
Outside of work, Rafał enjoys spending time with his family and helping his two sons explore the world.
Mateusz Wosiński
Machine Learning Tech Lead, deepsense.ai
Mateusz is a Lead Machine Learning Engineer specializing in building production-ready AI systems and leading agile, cross-functional teams. With deep expertise in computer vision, Generative AI, and MLOps, he focuses on transforming research prototypes into scalable, business-impacting solutions.
He has contributed to several high-impact projects, including developing an app automating operations in a major HR platform with web agentic system, creating a semi-deterministic voicebot that leverages LLMs for intent understanding and outputs contextually appropriate responses from a conversation tree or implementing an agentic system capable of autonomously resolving Jira and Git-based developer tasks using state-of-the-art LLMs.
Mateusz holds a Master’s degree in Applied Mathematics from the Technical University of Łódź and a postgraduate degree in Deep Learning from the Warsaw University of Technology.
