A story about discovery, surprise, and the future of human-AI collaboration
No existing code. No documentation. No hand-holding.
Just a test tenant and a simple instruction.
When the user provided different permissions than expected, Amplifier didn't ask what to do. It pivoted.
Participants just followed instructions Amplifier gave them.
Navigate to Azure portal. Grant these permissions. Click here. Paste this.
Based on participant accounts
With working M365 integration, participants explored what's actually possible.
Not API consumers. Actual team members.
When Teams split "@Amplifier Agent 1" into 3 separate mentions, Amplifier diagnosed and fixed it without being asked.
Data as of: February 20, 2026
Feature status: Experimental
Research performed:
git log --oneline on ramparte/amplifier-bundle-m365-collab (9 commits)git log --oneline on amplifier-m365-platform (37 commits)find . -maxdepth 4 -type fgit log --format="%an" | sort | uniq -cgit log --all --grep="hackathon" (0 matches)Gaps: No formal "hackathon" event found in git history. Narrative based on internal exploration sessions, January 2026. Success rate (100%) is from participant accounts, not formal measurement. Channel count (8) not independently verified.
Primary contributors: Sam Schillace (platform + collab), Marc Goodner (platform PRs)
The bottleneck isn't the AI's capability.
It's our imagination about what to ask for.