

AI-Native Development: How to Stop LLMs From Producing Code Slop
The proven framework for de-slopping your AI development.
This session introduces the Deslopify Methodology, a practical framework for guiding AI through executable tests, small steps, and modular architecture.
The goal is simple: help AI produce code you can trust.
What You Will Learn
• How TDD acts as an anchor for AI
• How small steps limit hallucinations and keep changes reversible
• How modular design helps the model reason within safe boundaries
• Spec driven development and CLAUDE.md ground rules
About the Instructor
Tsvetan Tsvetanov is a Lead Software Engineer and AI-native product builder with twelve years of experience delivering systems across gaming, transit, enterprise search, and clinical software.
He combines product thinking, test driven development, and modern AI tools to help teams ship high quality software quickly while minimising technical debt. He has worked with Via, Akooda, Flymingo, and Camplight, mentoring teams on adopting AI responsibly without losing engineering discipline.
Agenda
16:00 to 16:05
Welcome and event kickoff
16:05 to 16:45
Hands-on session with Tsvetan
How TDD, small steps, and hexagonal (modular) architecture enable clean, reliable AI-assisted development.
16:45 to 17:00
Live walkthrough and implementation patterns
Practical examples of steering AI and eliminating code slop.
Q&A and closing takeaways
Built by Builders, Backed by the Best
We’ve hosted engineers, financial experts, founders, and researchers from some of the world’s most forward-thinking companies. Here’s a look at who’s been in the room.
The last session filled up fast, save your spot before registrations close.