Team workshops

Workshop on Personal Productivity with AI

Private cohort workshop for managers, directors, and high-performing ICs. These pages spell out format, sample exercises, tooling, logistics, optional post-workshop engineering, and how to book.

All team workshops About me

Glossary

Assistants, APIs, and the words IT uses

Claude: Anthropic assistants and APIs for drafting and analysis. In productivity workshops we use Claude for structured passes over threads or specs when policy allows redacted or synthetic inputs, not as silent autopilot on external send.
Microsoft Copilot: Microsoft 365 embedded assistant in Word, Outlook, Teams, and related surfaces. Fits teams where daily work already lives in your Microsoft tenant with known retention.
Codex: OpenAI coding-oriented assistant inside editor flows. Shows up when technical leads or ops maintain small scripts, parsers, or automation next to team prompt packs.
Claude Code: Anthropic coding agent across a repo from terminal or editor. Same guardrails as other coding agents: strong for playbook files and glue code, not a substitute for policy review on customer-facing text.
ChatGPT: OpenAI chat products, including enterprise workspaces when enabled. Common first stop for drafts and variants. Still needs your written rules before anything sensitive or external ships.
Gemini: Google assistant and Workspace integrations. Fits when mail and docs already live in Google and admins allow the paste classes you use in exercises.
Cursor: IDE-first assistant with diffs and project context. Useful when teams version prompts, checklists, and short automation next to text instead of hiding them in chat history.
API: A machine-readable contract between systems over HTTPS with scoped keys. We point to where APIs help with calendar, docs, or ticketing, and where they should stay read-only for desk workers.
Integration: Wiring approved tools so summaries or drafts land where work belongs: CRM, ticketing, or shared drives. Different from pasting exports into a personal chat tab IT cannot see.
Webhook: HTTPS callback when an event fires, for example a form submit or ticket update. Common glue when teams graduate from copy-paste into light automation with owners.
LLM (large language model): The model behind assistants: next-token prediction from your prompt and context. It does not know your private backlog unless you connect it through approved paths or paste policy-safe snippets.
Shadow IT: Unapproved tools that spread because the sanctioned path feels slow. The workshop names the risk and pairs it with a faster approved alternative managers can sponsor.
Human-in-the-loop: Named reviewers before external send, before customer-visible edits, or before identifiers leave approved surfaces. Written as checklist steps with owners.
PII minimization: Share only fields the task needs, redact in classrooms, and route regulated data through systems built for it. Applies to HR notes, health, financial, and customer identifiers alike.
Model drift: Behavior shifts when providers ship new versions. Mitigation: pinned prompts, dated playbook reviews, and spot checks on high-risk templates after upgrades.