blog
Notes from the studio.
How Frenchie works, what we picked and why, how MCP is quietly reshaping agent tooling. No fluff. RSS available.
Read, listen, create: why your AI agent needs all three
Your agent already reasons. Give it three hands to do work: read files, listen to recordings, create images. Here's how image generation completes the Frenchie toolkit — and what that means for the shape of your next agent workflow.
Read the post →Why we built Frenchie: the MCP tools gap
Agents can write code, reason about systems, and orchestrate workflows. But drop a scanned PDF or a 30-minute recording in front of one and most of them hit a wall. That wall is why Frenchie exists.
Read the post →How Frenchie handles 30-minute audio without blocking your agent
Sync transcription is a great way to freeze an agent conversation for five minutes. Async job handling is the small engineering detail that makes transcription actually usable inside a live agent workflow.
Read the post →Inside Frenchie: from scanned PDF to clean Markdown in 3 seconds
What actually happens between dropping a scanned contract into your agent and getting back searchable Markdown with preserved tables and figures. A walk through the OCR pipeline's five stages.
Read the post →MCP for beginners: what it is and why Frenchie picks it
The Model Context Protocol is the quiet standard reshaping how AI agents get tools. Here's what it actually is, why it matters more than most people realize, and why Frenchie bet the whole product on it.
Read the post →Pay-per-use vs subscription: why Frenchie ships flat $1 = 100 credits
Every SaaS company eventually debates usage-based vs subscription pricing. For a narrow tool like Frenchie, the answer was clear before we wrote the first line of billing code. Here's the reasoning.
Read the post →
The best way to know us is to try what we build.
100 free credits, no card. Read a post, drop a file, see if Frenchie earns a spot in your stack.