​AI Agents in Real Workflows

​What changes when AI systems move from generating output to taking action

​Tuesday, April 28th, 2026
6:00–8:30 PM
Austin, TX (private venue shared upon confirmation)

​Presented in partnership with Tecla
Helping AI-forward companies hire elite talent across product, engineering, and growth.

​Food and drinks for the evening are generously provided by our partner.

​Invitation-only. Curated for senior technical leaders and operators.

​About the Event

​AI agents are moving from demos to production.

​In demos, they generate output.
In production, they take action.

​They trigger workflows. Modify data. Communicate with customers. Act across APIs and internal systems.

​When they misfire, the consequences are operational, not theoretical.

​This roundtable focuses on what actually happens when agents touch real systems: where risk surfaces, how autonomy is constrained, and who owns failure when something goes wrong.

​No speculation. No hype.
Just technical leaders comparing live deployment lessons under real accountability.

​The Curated Circuit Format

​GILD is intentionally small and highly curated.

​This is a private, off-the-record exchange under Chatham House Rules.

No panels. No pitches. No demos.

​After a short framing segment, the room moves into rotating micro-roundtables. Every table discusses the same prompt each round. Groups rotate so you meet a meaningful cross-section of the room.

​Designed for candor. Built for operators.

​Timed Agenda

​6:00–6:30 PM – Check in, Food & Drinks
6:30–6:45 PM – GILD welcome + sponsor introduction
6:45–7:05 PM – Framing discussion
7:10–8:05 PM – 3 breakout rounds (15–20 minutes each)
8:05–8:30 PM – Open mingling

​Speaker Framing & Friction

​The speaker will create productive tension around:

• ​Demo performance vs. production behavior

• ​False confidence in “human-in-the-loop” safeguards

• ​Expanding risk surface area as agents gain autonomy

• ​Where governance is lagging in deployment
  

​Breakout Prompts

​Round 1 – Reality Check

​When AI agents in your organization took real action in production, what actually happened, and where did unintended consequences surface?

​Round 2 – Failure Modes

​How are you deciding how much autonomy to grant agents and where have you pulled that autonomy back after something broke?

​Round 3 – Accountability & Containment

​When an AI agent makes a wrong decision that affects customers, revenue, or operations, who is accountable and is your organization structured for that reality?

​If you scaled agents 10x from here, what would fail first?

​Who This Is For

​Reserved for senior technical leaders with real authority over architecture, deployment, and operational risk:

• ​CTOs

• ​Technical founders

• ​VP / Head of Engineering

• ​Senior platform leaders

​Attendance is limited to protect candor and room quality.

​Privacy

​Chatham House Rules apply.
Discussion stays private.