Day 4 - Agentic Engineering/VIbe Coding an AI App from Zero To Production | 🔴 Live

2026-05-10 · Source: Venelin Valkov · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Software Development & Engineering, Robotics & Autonomous Systems · Depth: Intermediate, extended

Summary

This live stream details the ongoing development of an AI application using a combination of GPT 5.5 and Opus 4.7 for code generation and design. The application is deployed on Vercel, utilizes Clerk for authentication supporting Google, GitHub, and custom email, and is considering Convex for its backend database, noting its integration with Next.js and AI agent components. The design system, largely AI-generated, supports dark and light themes across six color palettes, with specific fonts like Spectral for titles and Inter for body text, and Phosphor icons for habits. Testing is set up with Biome for linting, Vitest for unit tests, and Playwright for end-to-end tests, with Convex's in-memory database testing highlighted. The current implementation includes user and category tables, with a focus on building out core features like goals and habits, demonstrating an AI-assisted workflow for issue implementation and design refinement.

Key takeaway

For AI Engineers building full-stack applications, integrating advanced LLMs like GPT 5.5 and Opus 4.7 into your development workflow can significantly accelerate feature implementation and design. Focus on providing clear design specifications and leveraging AI agents for both code generation and visual verification to maintain consistency and efficiency, especially when dealing with complex UI components and backend logic. Your testing setup should be robust, incorporating unit, end-to-end, and database-specific tests.

Key insights

AI-assisted development streamlines application building, from backend integration to UI design and testing.

Principles

• Combine LLMs for diverse development tasks.
• Prioritize robust testing from the outset.
• Iterative AI-driven design refines user experience.

Method

The workflow involves referencing PRD and design specs, using AI agents (GPT 5.5, Opus 4.7) with browser and frontend design skills for code generation, visual verification, and iterative design refinement, followed by comprehensive testing.

In practice

• Deploy Next.js apps on Vercel's free tier.
• Use Clerk for flexible user authentication.
• Explore Convex for Next.js backend and AI agent integration.

Topics

• AI Agentic Development
• Next.js Application Stack
• Convex Database
• AI-Generated UI/UX
• Automated Testing

Best for: AI Engineer, Machine Learning Engineer, Software Engineer

Related on AIssential

• Day 5 - Agentic Engineering/VIbe Coding an AI App from Zero To Production | 🔴 Live — AI agents can achieve high test coverage and implement features in complex applications.
• Dispatches from O'Reilly: The accidental orchestrator — Disciplined AI-driven development requires human orchestration of AI agents, utilizing structured practices and critical thinking for effective outcomes.
• A new era for software testing — LLMs can significantly enhance software QA and testing by automating complex, traditionally manual processes without compromising quality.
• [FULL WORKSHOP] AI Coding For Real Engineers - Matt Pocock, AI Hero (@mattpocockuk ) — Software engineering fundamentals, like small tasks and clear feedback, are crucial for effective AI integration.
• Day 1 - Agentic Engineering/VIbe Coding an AI App from Zero To Production | 🔴 Live — AI-driven interview skills can efficiently flesh out complex application specifications and product requirements.
• Spec-Driven Development: AI Assisted Coding Explained — Spec-driven development provides a structured, SDLC-aligned approach to AI-assisted coding, reducing ambiguity for LLMs.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Venelin Valkov.