xpander.ai’s new step-by-step system makes AI agent more reliable

The Agent Graph System (AGS) from Israeli startup xpander.ai represents a significant advancement in making AI agents more reliable and efficient when handling complex, multi-step tasks.

Core innovation: xpander.ai’s Agent Graph System introduces a structured, graph-based workflow that guides AI agents through API calls in a systematic manner, dramatically improving their reliability and efficiency.

• The system restricts available tools at each step to only those relevant to the current task context, reducing errors and conflicting function calls
• AGS works with underlying AI models like GPT-4 to enable more precise automation workflows
• The technology includes AI-ready connectors that integrate with systems like NVIDIA NIM, enriching API tools with detailed documentation and schemas

Performance metrics: Benchmarking tests demonstrate substantial improvements in AI agent performance when using AGS combined with Agentic Interfaces.

• AI agents achieved a 98% success rate in multi-step tasks, compared to 24% using traditional methods
• Workflows were completed 38% faster than conventional approaches
• The system used 31.5% fewer tokens, leading to reduced operational costs

Technical leadership: The founding team brings significant enterprise technology experience to address common AI agent challenges.

• CEO David Twizer and CPO Ran Sheinberg both previously served as principal solutions architects at Amazon Web Services
• Their experience with large-scale enterprise computing informed AGS’s design to be both powerful and accessible
• The team focused on creating technology that can integrate with existing systems while allowing for future model upgrades

Real-world applications: The system has demonstrated practical benefits in complex business workflows.

• One benchmark test involved coordinating research across LinkedIn and Crunchbase, with results organized in Notion
• The system ensures tools are used in the correct sequence while maintaining consistent schema compliance
• AGS includes built-in error management and fallback options, allowing agents to retry failed operations or find alternative workflows automatically

Future implications: As AI agents become more prevalent in enterprise environments, AGS’s structured approach to handling complex workflows positions it as a key enabling technology for the broader adoption of automation solutions.

• The system’s ability to manage error handling and maintain context continuity addresses critical challenges in enterprise AI deployment
• By making AI agent development more accessible, AGS could accelerate the adoption of automation across industries
• The focus on API integration and structured workflows suggests a future where AI agents can reliably handle increasingly complex business processes