Send AI
Amsterdam-based IDP provider using custom open-source AI models for document automation in logistics, insurance, and government sectors.
Overview
Send AI differentiates through self-learning AI models that customers train on their specific data, contrasting with the commercial model approach used by most IDP vendors. Founded in Amsterdam in 2021 as Autopilot by Thom Trentelman and Philip Weijschede, the company raised €2.2M in pre-seed funding from Google's AI fund. The platform processes complex documents including low-quality scans and photographed paperwork using combinations of smaller open-source models rather than large commercial alternatives.
However, Send AI faces visibility challenges in an increasingly competitive market. Gartner Peer Insights tracks 93 IDP solutions with Amazon Comprehend, ABBYY, and Nanonets leading customer recommendations, while Send AI was absent from analyst coverage in early 2026. The company also confronts pricing pressure from new entrants like Mistral AI's OCR 3 at $2 per 1,000 pages.
Key Features and Benefits
- Self-Learning AI Models: Custom model training on organization-specific data with isolated processing
- Open-Source Model Architecture: Uses combinations of smaller open-source models instead of commercial alternatives
- Document Pre-Processing: Automated preparation and segmentation for complex files
- Human-in-the-Loop Validation: Manual verification workflows for extraction accuracy
- Custom Rule Engine: Configurable validation rules for data quality assurance
- Data Ownership: Full customer control of data and trained models
Use Cases
Logistics Document Processing
Processing shipping documents, bills of lading, and customs paperwork from photographed documents and low-quality scans with custom models trained on company-specific formats.
Insurance Claims Automation
Automating claims processing from diverse document sources, handling lengthy PDFs and applying custom validation rules while maintaining GDPR compliance.
Government Document Management
Processing citizen submissions, forms, and administrative documents with modular segmentation and human-in-the-loop verification workflows.
Technical Specifications
| Feature | Specification |
|---|---|
| AI Technology | Self-learning models, open-source model combinations |
| Model Approach | Custom, isolated models per organization |
| Document Types | Low-quality scans, photographed documents, lengthy PDFs |
| Processing | Modular segmentation, pre-processing automation |
| Validation | Custom rules, human-in-the-loop |
| Infrastructure | Google Cloud Platform |
| Compliance | ISO27001:2022, GDPR, DORA |
| Deployment | Cloud-based, scalable |
| Data Ownership | Full customer ownership of data and models |
| Target Industries | Logistics, insurance, government, maritime, finance, retail |
Resources
Company Information
Headquarters: Amsterdam, Netherlands
Founded: 2021 (originally as Autopilot)
Founders: Thom Trentelman, Philip Weijschede
Funding: €2.2M pre-seed round (Google's AI fund)