The problem
AI products misread tables, forms and messy layouts, so document-heavy workflows can't be trusted or automated
What they do
API and no-code studio that turns PDFs, scans and spreadsheets into structured data for AI
Traction $10M+ ARR (est. ~$20M); revenue up 8x YoY; billions of pages processed
Hiring trend
- Open roles
- 36
- Ops roles
- 4
- Headcount
- ~70-95
Team +91% in 12 mo (Dealroom); ~79 on LinkedIn; 40 open roles, sales build-out since Aug 2026
Hiring in
- San Francisco
- New York
- Remote (North America)
Funding
- Last round
- $75M
- Stage
- Series B
- Date
- Oct 2025
Led by Andreessen Horowitz, Benchmark
- Hiring+236 open roles on the live job board (30+ roles = +2)
- Funding recency+1Series B · $75M, Oct 2025 (12 months ago): 6–12 months = +1
- Traction+1.5Strong growth: $10M+ ARR (est. ~$20M); revenue up 8x YoY; billions of pages processed
- Red flags0None found in the last 24 months
- Total4.52 + 1 + 1.5 − 0 = 4.5 → Race (4.25–4.74)
Distance (Half) is a separate read on how long the lead can last. It doesn't change the score.
Distance
Half · edge: hard to leave
Leading AI-native document parser (Harvey, Scale, Vanta, Toast, a Fortune 10) with its own r-1 model and a widening workflow suite; still a crowded field (LlamaParse, Unstructured, Google/Azure) and labs like Mistral sell cheaper OCR straight to developers.
Sources: Deep Analysis: fastest-growing IDP startup, $10M+ ARR (May 2026) ↗ · Revenue Memo: Reducto funding, customers and risks (Oct 2026) ↗ · PR Newswire: r-1 parsing model launch (Sep 2026) ↗
Last checked Oct 25, 2026