Mushegh Gevorgyan is a software architect, founder, and engineering leader with more than fifteen years of experience designing and building distributed systems, developer platforms, and AI-driven products. His background spans hands-on software development, system architecture, and technical leadership across startups and enterprise environments.

He is the CEO and Founder of Dowork Technologies, where he has led the development of AI-powered voice agents and workflow systems used in production to automate complex, real-world processes. His work emphasizes reliability, observability, and explicit system design in environments where AI components are deeply integrated into core business logic.  

Earlier in his career, Mushegh served as a principal developer and software architect, working extensively with backend systems, distributed services, and large-scale application design. Across roles, he has consistently focused on reducing system complexity through clear abstractions, well-defined workflows, and strong architectural boundaries.

In recent years, his work has centered on the challenges of AI-assisted software development, specifically how unstructured use of AI can introduce hidden coupling, architectural entropy, and long-term maintenance risks. He advocates for disciplined, architecture-first approaches that allow teams to benefit from AI productivity gains while preserving system integrity, correctness, and human oversight.

Presentations

23x

Beyond Vibe Coding: How to Scale AI-Assisted Development Without Architectural Chaos

AI coding assistants have made engineers 10x faster, but when entire teams use them, a new problem emerges: architectural chaos. Each developer gets working code from their AI, but without shared architectural context, you end up with conflicting patterns, duplicate services, and unmaintainable systems. This talk introduces SpecMind, an open-source tool that enables spec-driven vibe coding by automatically analyzing your codebase, generating architecture documentation, and keeping design decisions aligned across your team. Learn how to maintain architectural consistency while moving fast with AI, and see a live demo of the analyze, design, and implement workflow that keeps teams building together instead of in isolation.

See Presentation