CTO panel: Internal apps and agents in production
summary
Agents analyze recorded customer calls and auto-generate code and pull requests, marketers without GitHub use a starter kit (Claude Code, Vercel, GitHub SSO) in an AI accelerator for non-engineers, and one org drops Tableau and replaces BI with a v0-powered control plane that handles spreadsheets and Snowflake. Ephemeral Vercel previews let non-engineers iterate; CI/CD plus CodeRabbit, evals and custom compliance models catch vulnerabilities and Fair Housing issues. Some projects stay demo-only, others scale into production for thousands of users.
AI-written summary · about
chapters (8)
- 0:00Meet the panel
- 0:57What seemed unrealistic a year ago
- 3:40Who gets to build: MJH's AI accelerator
- 4:47Vertical AI for real estate at SERHANT
- 6:24ZoomInfo + v0: meeting builders on a spectrum
- 7:39Security, guardrails, and the shadow IT battle
- 12:45Agent workflows that mirror human teams
- 18:23If an agent could do one thing today
related talks
27mCTO Panel: Agents in production
24mGitHub Copilot Agents + Sonar | The Blueprint for Agentic Development
50mPanel: Beyond The Vibe: Code Quality First
25mFrom Analytics to Agentic Apps
18mFrom POC to Profit: Deploying Customer-Facing AI Agents That Actually Work
12mHow to maintain a codebase – when everyone can code
32mBeyond the Hype: A Pragmatic Look at How Developers Actually Use AI Tools
26mAI Agents as Your Co-Developer