Hoshii AI Assistant Doubles Order Processing

💡See how an ERP-connected AI assistant could double order-processing speed.
⚡ 30-Second TL;DR
What Changed
Hoshii is developing an AI assistant for ERP-connected workflows.
Why It Matters
ERP-connected AI assistants could reduce manual order-entry work and improve operational throughput. Enterprises will still need to validate accuracy, exception handling, and integration costs before deployment.
What To Do Next
Prototype an ERP order-processing workflow with structured tool calls and measure accuracy, latency, and exception-handling rates against manual processing.
Key Points
- •Hoshii is developing an AI assistant for ERP-connected workflows.
- •The assistant targets faster and more efficient order processing.
- •Reported processing speed is expected to reach twice the previous level.
🧠 Deep Insight
AI-generated analysis for this event.
🔑 Enhanced Key Takeaways
- •Hoshii is a Swiss-based technology firm specializing in autonomous business process automation, specifically targeting the integration gap between legacy ERP systems and modern generative AI.
- •The AI assistant utilizes a proprietary 'Human-in-the-Loop' (HITL) verification layer that allows ERP data to be validated by human operators before final transaction execution.
- •The system is built to support multi-language order processing, enabling Swiss enterprises to handle cross-border trade documentation automatically within their ERP environments.
- •Hoshii's architecture focuses on reducing 'ERP fatigue' by automating repetitive data entry tasks that typically account for 60% of administrative order-processing time.
- •The solution is currently being deployed as a middleware layer, ensuring compatibility with major enterprise ERP platforms like SAP and Microsoft Dynamics 365.
📊 Competitor Analysis▸ Show
| Feature | Hoshii AI | UiPath (Automation Cloud) | Celonis (Process Mining) |
|---|---|---|---|
| Primary Focus | ERP-native AI Assistant | Robotic Process Automation | Process Optimization/Mining |
| Integration | Direct ERP API/UI | Broad Enterprise Ecosystem | Data-driven Process Analysis |
| Speed Impact | 2x Order Processing | Variable (Task-based) | Indirect (Insight-based) |
| Pricing Model | Usage-based/Subscription | Tiered Licensing | Enterprise/Volume-based |
🛠️ Technical Deep Dive
- Utilizes a Retrieval-Augmented Generation (RAG) framework to query specific ERP database schemas in real-time.
- Implements a secure API gateway to ensure data privacy and compliance with Swiss data protection regulations (FADP).
- Employs Large Language Models (LLMs) fine-tuned on structured business document formats (invoices, purchase orders, shipping manifests).
- Features an asynchronous processing engine that allows the AI to queue tasks during high-volume periods without locking ERP user sessions.
🔮 Future ImplicationsAI analysis grounded in cited sources
⏳ Timeline
Weekly AI Recap
Read this week's curated digest of top AI events →
👉Related Updates
AI-curated news aggregator. All content rights belong to original publishers.
Original source: 钛媒体 ↗



