Test the investment case against the data room. Trace revenue quality, debt terms, and downside assumptions into a cited IC draft.
More research capacity. Every conclusion traceable. Give Analyst One the mandate, the source set, and your investment context. It turns them into cited analysis, reports and workbooks.
The system does the research work. Your team retains judgment and final approval.
It’s the foundation. Every answer, every figure, every summary traces back to a line in a document.
Our ingestion, OCR, and entity layer are our own. We control the primitives so the quality compounds. No shortcuts.
Analyst One behaves like a seasoned analyst. Not standing in your way, but delivering precise, skeptical, and accurate work.
Test the investment case against the data room. Trace revenue quality, debt terms, and downside assumptions into a cited IC draft.
Reconcile definitions across the credit agreement and amendments, track leverage and headroom, and cite the governing clause in the memo.
Work across filings, transcripts, and external data to explain what changed — and draft the earnings note with every figure traced.
Seat-based, with volume discounts from five seats. Until then, tell us about your desk and we'll walk through what fits.
Pages travel to our servers for OCR and extraction — briefly, under contract, and never to train a model. The parsed index lives on your device, and your queries against it run there too.
For anything not covered here, email hello@felixresearch.com. James reads everything.
September 1. Access is strictly gated. We review early-access requests and onboard firms directly. Leave your work email below — no account is needed until access is approved.
Give it an objective and it works through a bounded research loop: it plans what to investigate, searches the approved source set, reads the relevant pages, and drafts a persistent deliverable. Retrieval is read-only; proposed changes to existing work stay pending until you accept them.
PDFs (text or scanned), plain text, HTML, and links — paste a single SEC URL or an entire investor-relations site. Scanned PDFs are OCR'd automatically.
Pages travel to our EU-hosted infrastructure for OCR and extraction, and the temporary upload is auto-deleted — within 24 hours at most. The parsed index is stored on your device. Your data is never used to train a model.
Seat-based, with volume discounts from five seats. The full schedule goes public at launch on September 1 — until then, contact us and we'll walk through what fits your desk.
Leave your work email. We review every request and onboard firms directly. If there is a fit, we'll get in touch about access.