Compound launches IC Deck Bench, first benchmark for IC-ready PowerPoint agents
peterjliu · x · 2026-11-17
Compound released IC Deck Bench, the first benchmark measuring how well AI agents produce investment-committee-ready PowerPoint decks for private equity, spanning 61 tasks (data room to IC deck, lender decks, firm templates, post-committee updates, exhibits, edits). Scoring targets agent+model+harness combos. Compound's own harness scores 89.2% at $4.38/task, 11 points above the best other agents at that cost. Motivated by the gap between frontier agents and an intern-quality deck.
Related event: Compound Launches IC Deck Bench for AI-Generated Investor Decks(2 posts)→
More from coding & agent
- After a compromise, this engineer moved all AI coding agents off his desktop — rseroter · 2026-11-17
- Confluent's Kafka Agent Skills Praised for Quality, Ship With Built-in Evals — AdiPolak · 2026-11-17
- itr8 puts Claude, GPT, Gemini and Grok in one group chat to collaborate on your work — cantrell · 2026-11-17
- 1000-agent swarm pybot experiment writeup coming within days — lvwerra · 2026-11-17
- Developer: My Elaborate AI Workflows Kept Failing — Just Focus on the Outputs — brandon_galang · 2026-11-17
- Gemini CLI Fix Stops Custom Model Aliases From Silently Losing Generation Settings — sayaklearner-prog · 2026-11-17