

AI-Agentic Development Workflow
For longer-running projects, AI-assisted development needs a repeatable workflow: how work is planned, what context is loaded, what gets documented, how progress is reviewed, and how the next session continues without starting from zero. The recurring challenges are context management, documentation, and orientation. Maintaining quality as projects grow requires treating the workflow itself as an engineering problem.
About the speaker
Vardan Arakelyan is a Software Engineer at AAS Devel team with a background spanning software development and systems architecture. His practice with AI-assisted development is grounded in real engineering environments where workflow discipline and context continuity aren't optional extras, they're the difference between AI being useful and being noise.
Who is this for?
Developers with hands-on experience; juniors who are already building, mids and seniors looking to systematize their AI usage. If you've worked with AI coding tools and want to take it to the next level, this is for you.
What you'll take away:
A repeatable workflow structure for AI-assisted development
Strategies for context management and documentation
How to maintain code quality and project orientation across sessions
Real patterns from production-level use, not toy examples.
The session will be held in Armenian.