Typewise Launches Multi-Agent Orchestration for Enterprises
AI company Typewise has introduced a multi-agent orchestration platform for enterprise customer service. The system uses an AI Supervisor Engine to coordinate specialized AI agents, built with natural language, to resolve complex cases. The platform is designed to manage human handoffs to achieve faster and more accurate resolutions in production environments.
- Zurich-based Typewise was founded in 2019 by CEO David Eberle and CTO Janis Berneker, initially launching a privacy-focused consumer keyboard app that was downloaded over 1.5 million times before the company pivoted to B2B enterprise solutions after participating in Y Combinator's Summer 2022 batch. - The platform's architecture, where an AI Supervisor routes tasks to specialized agents, is a practical implementation of the coordinator pattern in multi-agent systems, designed to handle complex workflows that a single agent cannot. - A key feature for platform teams is the use of natural language for creating and modifying automation workflows, allowing business users to describe outcomes directly without needing to code or design flowcharts. - For the logistics sector, DPD (Schweiz) AG used Typewise's AI to increase customer service efficiency by more than 25%, a project that was recognized as a 2024 European Customer Champion. - Typewise reports that clients see a 4-6x return on investment through reduced handling times and has integrations with over 200 connectors for enterprise systems like CRMs and ERPs. - The platform's approach to human-in-the-loop automation allows for granular control, enabling partial handoffs or requiring human approval for specific steps, ensuring that AI agents handle workflows while humans control the final outcomes. - The market for AI in customer service is projected to grow from $12.1 billion in 2024 to $47.8 billion by 2030; however, a 2025 survey indicated that only 12% of customer service executives had adopted fully autonomous AI agents for case resolution. - The company has raised a total of $4.61 million over four funding rounds and is backed by investors including Y Combinator, L-SPARK, and ACE Ventures.
