From Vibe Coding to Spec-Driven Development with AWOS in Claude Code

Vibe coding works great until it doesn't. When AI agents start ignoring your architecture, making wrong assumptions about your stack, and producing code that compiles but misses the point, the problem isn't the model. It's the instructions. This talk introduces AWOS (Agentic Workflow Operating System), an open-source framework built by Provectus for Claude Code that brings Spec-Driven Development to AI-assisted coding. AWOS structures the development process into 8 phases, each with its own specialized agent and audience: Product Definition, Roadmap, Architecture, Functional Spec, Technical Spec, Tasks, Implementation, and Verification. The result is a traceable path from a high-level idea to every line of code, with human checkpoints at every stage. - What you'll see: a live demo building a conference talk management app (yes, meta), comparing vibe coding versus the AWOS workflow side by side. - What you'll learn: why specifications are just really good context for LLMs, how idempotent document structures solve the context loss problem, and where Spec-Driven Development fits alongside tools like GitHub Spec Kit and AWS Kiro. - What you'll take home: a tool you can install with npx @provectusinc/awos and start using immediately.

Speaker

Elbano Mibelli P.

Elbano Mibelli P.

Senior DevOps Solutions Architect @ Provectus

Elbano is a Senior DevOps Solutions Architect at Provectus, an AWS Premier Consulting Partner and official Anthropic partner, with over 9 years of experience in cloud-native, AI, DevOps, and distributed systems design. He holds two AWS certifications, three AWS Black Belt specializations, and an ITIL Foundation credential. His career started in telecommunications engineering in Venezuela, moved through fintech integrations, full-stack development, and IT management in Colombia, and eventually converged on cloud architecture and AI. Before joining Provectus, he led a team of 17 solutions architects across 5 regional teams in LATAM and spearheaded a Generative AI practice that achieved AWS Generative AI Competency. Today, he works at the intersection of cloud infrastructure, agentic AI, and enterprise delivery, designing solutions that range from multi-account AWS platforms for AI agent orchestration to intelligent contact centers with Amazon Connect and Bedrock. The frustration of watching AI coding agents ignore architectural decisions and produce code that "sort of works" in real enterprise engagements is what drives this talk. As a daily Claude Code user, Elbano has experienced firsthand both the power and the limits of AI-assisted development, and believes the most valuable skill for the next generation of developers will not be writing code, but communicating intent with enough precision that an agent gets it right the first time.

View speaker

Want to know more?

Join PyCon Colombia newsletter and get a complete overview of our events, speakers and community participation.