Register or log in to access this video
New York • September 8 & 9, 2027
Loved LDX3 New York? Pre-sale tickets for 2027 are now available.
A practical look at the SDLC mechanics that separate teams getting real output from agents from teams generating a lot of code that doesn’t hold up. This talk covers spec-driven development as a way to give agents better constraints, large language model (LLM) gateway and routing strategies for managing which model handles which task, test-driven development adapted for an agent-first workflow, and verification practices built for the volume and pace of AI-generated code. This session is a deep dive on the actual mechanics of building software well with agents in the loop.
The talk also covers the part most teams haven’t started on: the environment itself. CI, version control, code review tooling and codebases were all designed for humans as the primary user. When agents become the primary user, that assumption quietly breaks. Agents don’t fail randomly — they fail when context is lost. Designing environments agents can navigate, verify against, and operate in safely and autonomously is where the next tranche of productivity actually lives.
You’ll leave with:
- A framework for writing specs that give agents better constraints to work within
- A way to think about LLM gateway and routing decisions, including when to route tasks to different models
- A test-driven approach adapted for an agent-first workflow
- A set of verification practices built for the volume and pace of AI-generated code
- A view of what an agent-ready engineering environment looks like — and where your current setup still assumes a human is driving
