Gen AI Leadership Forum · Agentic AI Bootcamp
Chatbots answer. Agents act.
Build production-grade AI agents in TypeScript — the ReAct loop, custom tools, a four-tier memory stack and multi-agent systems, in one hands-on day with three codebases you keep.
60–80%
reduction in manual repetitive work
24/7
autonomous continuous operation
0
production codebases to take home
0
learning blocks, zero skill gaps
[ Core Stack ]
[ From Chatbots to AI Agents ]
Wiring up an LLM call is easy. Shipping AI agents that use tools, remember across sessions and coordinate with each other is what this bootcamp builds.
>Where is my order, and can you change the delivery address?
✗ Chatbot
Explains how order tracking works. It cannot look the order up or change anything.
✓ AI agent
Checks the order through a tool, changes what it is allowed to change and escalates the rest to a person.
>Use the same format as last time.
✗ Chatbot
Starts from zero every session, so the preference has to be repeated.
✓ AI agent
Recalls the preference from its four-tier memory and answers in the saved format.
>Write a report on this market, with sources.
✗ Chatbot
Produces one draft in one pass, with nobody checking it.
✓ AI agent
A researcher gathers, a writer drafts and a reviewer sends it back for up to two revisions before it is done.
[ Who Should Attend ]
Software Engineers
Moving from single LLM calls to AI agents that use tools, keep memory and finish the task.
Tech Leads
Building AI agents for production, from the ReAct loop and state management to multi-agent patterns.
AI Engineers
Building support, research or personalisation agents that act on real systems instead of only answering.
Backend Engineers
Connecting agents to knowledge bases, APIs and services, and escalating what they should not decide alone.
Data Engineers
Designing agent memory across sessions on Redis and Pinecone, with importance scoring deciding what is kept.
AI Architects
Planning multi-agent systems in TypeScript where specialised agents coordinate on one research task.
[ The ReAct Loop ]
One loop separates AI agents from chatbots — reason about the next step, act through a tool, observe the result and update state, implemented in TypeScript.
Decide what the task needs next from the goal, the current state and the conversation so far.
Call a tool or hand the work to another agent.
Read the result back into the loop as it streams in.
Keep what matters and trim what does not, so the context stays efficient.
Stop when the goal is met, otherwise go round the loop again.
[ Tool Integration ]
Design and register custom tools that connect AI agents to external APIs, databases and services with LangGraph — validated with Zod, streamed, and run in parallel when the task allows.
[ Memory Systems ]
Build AI agents that remember user preferences across sessions with a four-tier memory stack on Redis and Pinecone, with importance scoring deciding what is kept.
The current exchange, kept in context
Recent sessions, fast to read
Long-term memories found by meaning
Decides what is worth keeping for next time
[ AI Agents Agenda & Curriculum ]
Each block builds directly on the last, from the agent loop to multi-agent systems in production, then the labs.
Learn how AI agents differ from chatbots, implement the ReAct loop and manage agent state in TypeScript.
Design and register custom tools that connect agents to external APIs, databases and services using LangGraph.
Build agents that remember user preferences across sessions using a four-tier memory stack with Redis and Pinecone.
Orchestrate specialised multi-agent systems with resilient communication, error handling and production monitoring.
Build the three deliverables: a customer support agent, a multi-tiered memory system and a multi-agent research system.
Leave with the key takeaways and the resources to keep building.
[ Three Codebases You Keep ]
Each lab runs about 25 minutes and ends with working AI agents in TypeScript — templates you can adapt to real business problems from day one.
Autonomous support
Personalised responses
Autonomous reports
[ Multi-Agent & Production Patterns ]
Orchestrate specialised AI agents in hierarchical, pipeline and peer patterns, with circuit breakers, observability and cost tracking so the system holds up in production.
Messages move between agents in the pattern you pick.
[ Your Instructor ]
Instructor
Sr. Software Engineer, ENTERSOFTONE
Alexandros is a Senior Software Engineer with 15 years of versatile experience in full-stack development. He specializes in C#/.NET, React.js, Python, WebGL, and AI development, including agent-based systems, RAG workflows, and Mastra AI SDK integration.
He also brings strong DevOps expertise with Azure, enabling scalable and secure deployments. A creative problem solver, he combines deep technical knowledge with intuitive UX/UI design, consistently staying ahead of industry trends to deliver modern, engaging applications.
[ Questions Engineers Ask ]
No. Every lab is written in TypeScript on Node.js.
Intermediate to advanced. The day starts from the agent loop and moves quickly into tools, memory and multi-agent systems.
Three working codebases — a customer support agent, a multi-tiered memory system and a multi-agent research system.
A laptop with Node.js and TypeScript, and your API credentials. Setup runs during registration at 09:30.
October 22, 2026, at Telekom Academy in Athens, from registration at 09:30 to the wrap-up at 18:15, as part of the Gen AI Leadership Forum.
It is one instructor-led day in the room, building with the instructor, and you leave with three working codebases instead of a set of videos.
Reserve your seat on the Gen AI Leadership Forum tickets page.
[ More Forum Bootcamps ]
Engineering · Oct 21
One day building knowledge systems that hold up in production — five blocks, and three codebases you take home.
Learn More →Product & Operations · Oct 22
One day that gives everyone one to three hours back, every day — no code, just the work rebuilt around AI.
Learn More →Leadership · Oct 22
One day for the people who set direction — seven modules across four strategic phases, no code.
Learn More →Eight hours, four blocks and three production codebases in TypeScript — October 22, 2026, at Telekom Academy in Athens.