Kore.ai launches Autoloop to keep tuning enterprise AI agents after they go live
Enterprise artificial intelligence platform company Kore.ai Inc. today launched Autoloop, an optimization engine for the AI agents that customers build on its Kore.ai Agent Platform. Businesses set the targets, and Autoloop keeps adjusting the agents to hit them automatically, including after they’re deployed. Autoloop goes after the way most enterprises maintain agents today, fixing failures […]…
Kore.ai Inc., an enterprise artificial intelligence platform company, recently introduced Autoloop, a tuning engine designed to optimize AI agents after they have been deployed. The optimization process is automated, ensuring that the agents maintain target performance levels, even after launch. Unlike traditional methods of fixing issues manually, Autoloop focuses on continuous improvement, addressing problems proactively.
According to Kore.ai's 2026 Kore.ai Agent Productivity Index, 79% of enterprises reported reverting actions taken by AI agents, while 70% faced untraceable failures. By setting goals related to task completion, adherence to business rules, cost, and accuracy, customers can leverage Autoloop to ensure their AI agents perform optimally.
Accuracy is assessed based on whether the answers are supported by the company's own data. Each proposed change is evaluated against the entire set of goals, preventing the compromise of safety or the agent's ability to complete tasks. Autoloop generates the initial version of each agent using existing operating procedures and iterates through multiple rounds of optimization in production.
Kore.ai's StateTrace evaluation layer identifies when a goal is missed, analyzing each interaction's details, including handoffs, tool calls, and state changes. A five-layer validation architecture supports round-the-clock optimization while keeping costs manageable. The Agent Blueprint Language, introduced earlier this year, simplifies the process by compiling routing, business rules, and guardrails into an executable state machine.
With this, Autoloop can rewrite only the specific piece responsible for a failure, enhancing precision. According to Kore.ai's CTO and CPO, Prasanna Arikala, these technologies make automatic optimization feasible. The company's approach also influences its software development, with AI agents contributing to about 6,500 commits per month in the company's 2.6 million-line codebase, adhering to 68 guardrails.
Founder and CEO Raj Koneru emphasizes that companies scaling AI will rely on AI to build, govern, and optimize AI. Autoloop is now accessible to all customers on Kore.ai Agent Platform's Artemis edition, which launched in May, and is utilized by over 500 Global 2000 organizations.
Rewritten by urgent.news; may contain errors — check the original.