AI Workflow Workshops
Two hands-on, 3-hour live workshops. One for the individual building their own productivity agent. One for the engineering team taking agents to production. Hosted by eCornell.
Why AI workflows, not just AI models
"The winners won't be the ones who use AI. They'll be the ones who redesign work around it." Better workflows beat bigger models. Reliability comes from systems: atomize tasks → constrain outputs → verify → iterate. The biggest performance gains aren't model upgrades, they're better workflows.
These workshops are the practical version of that thesis. Bring a real problem; leave with a working pattern.
Hands-On AI Agents, Vibe-Code AI Workflows for Personal Productivity
Vibe-code your own personal AI agent in three hours. What works, what doesn't, what's safe.
For: Operators, founders, and individual contributors who want to automate their own workflows, no coding required.
You leave with:
- A working agent skill running against your real workflow (e.g. email triage).
- A clear mental model of what agents are and where they break.
- A playbook for safely extending agents into more of your work.
Building Enterprise-Ready AI Agents, From Prototypes to Production
Production-grade agent architecture with Claude Code, LangGraph, and the OWASP LLM Top 10. Three hours.
For: Engineering leaders, AI architects, and senior PMs taking agents past the demo stage.
You leave with:
- An architecture blueprint for a deployable agentic system.
- A working understanding of the OWASP LLM Top 10 threat model and how to mitigate each.
- Pattern templates for RAG, planning, and tool-use that survive enterprise security review.
Want it for your team?
Both workshops can be delivered as a private cohort for an organization (corporate, university, or fund). hello@lutzfinger.com.