Launching today

Farside Research
Deep research you can check, in the app, API or MCP
1 follower
Deep research you can check, in the app, API or MCP
1 follower
Ask a hard question and Farside searches in rounds, grades the sources and writes a cited answer. Every claim links to the passage behind it, open gaps stay visible, and Deep runs chase the loose threads. Use it in the web app, REST API or MCP.






Free Options
Launch Team

AWS Startup AdvisorPersonalized AWS guidance that meets you where you’re already building
Promoted
Hey Product Hunt, I’m Mike, maker of Farside Research.
I built this after one too many AI research answers that sounded great and fell apart the minute someone asked “where’s that number from?” If you’re putting a recommendation in front of a client or your boss, a fluent paragraph isn’t enough. You need to see the source, and you need to know what the research didn’t find.
So Farside works like this: you ask a question, it searches in rounds, grades what it reads, and writes a synthesis where every claim points back to the passage it came from. Gaps and contradictions get their own sections instead of getting papered over. Deep runs follow the loose threads further and pull the related work into a dossier.
What else is in there:
Lite answers in a few minutes, for when you just need context before a meeting
The same runs through the web app, a REST API and a hosted MCP server, so Codex, Claude, Cursor or any other MCP client can estimate the cost and kick off research on their own
Pipelines that re-run research on a schedule and only ping you when the answer actually moved
Want to see it before signing up? Here’s a real run, published exactly as it came out: https://farsideresearch.com/sample (177 sources read, 97 kept, and you can click through every claim).
No card to start. Sign up and you get 28 credits right away, enough for 2 quick reports and a few follow-up questions. When you want more, add a card to unlock the full 7-day trial with 250 credits, and cancel any time before it ends.
What I’d love feedback on:
Open the sample report. Is it obvious what you’d trust and what you’d double check?
If you’d plug research into an agent or your own product, what’s missing from the API or MCP tools?
What’s the first question you’d hand it?
I’ll be here all day answering everything.