AI Code Archaeology: How to Give an Agent a Map, Memory, and Checks
About the webinar
In a legacy repository, documentation almost always lags behind the code. When an AI coding agent enters such a system, it fills in missing context with assumptions. The result may look convincing but might change the wrong part of the system or break dependencies the agent didn’t see.
In the webinar, Олексій Попов will show how to prepare a context harness for the agent: a codebase map for a specific task, a memory of verified facts, and evidence gates that separate knowledge from assumptions.
Using a real legacy codebase as an example, we will analyze a controlled agentic loop — from finding the minimal change zone to stop conditions under which the agent should stop before writing or continuing code.
Participants will receive ready-to-use skills that can be applied in their own brownfield projects.
The webinar is held within the EUNITED Digital Hub project. Co-financed by the European Union. However, the views and opinions expressed are solely those of the authors and do not necessarily reflect those of the European Union or the European Health and Digital Executive Agency (HaDEA). Neither the European Union nor the granting authority can be held responsible for them. The EUNITED Digital Hub project received funding under the Digital Europe Programme under grant agreement GAP-101269491.
What we will cover
Speaker
Oleksii Popov
17+ років в інженерії — від розробника до CTO. Впроваджує ШІ в повний цикл розробки та будує AI-native інженерні команди.
- Ex-CTO у CHI Software — інженерна організація з 800 фахівців
- Ex-VP of Engineering у GenAI.Works — платформа оркестрації ШІ-агентів для спільноти з 12+ млн учасників
- Керував інженерними організаціями на 250–300 FTE у Customertimes і S-PRO
- Solution Architect в EPAM; Head of Software Engineering у fintech-компанії Kreditech (Гамбург)
- AWS Certified Solutions Architect – Professional
У SET University керує програмами GenAI and Agentic AI for Software Engineering та Solution Architecture.
Who will benefit from the webinar
Key insights you will gain
- How to prepare the context before the agent starts changing the code
- How to build a map only for the necessary part of the system
- How not to mix verified facts with assumptions
- How to maintain useful project memory without accumulating outdated context
- How to limit the change zone and stop the agent in time
With the support of
Participation
Give the AI coding agent verified context instead of room for guesswork.
Participation is free with prior registration. Registration is open until October 15, 2:00 PM.
