Among the many brilliant use cases and builds, one solution stood out for its ingenuity and technical ability to leverage all the different features from Kizen’s AgenticOS platform: The Grand Prize winner Disha Anil, who built the use case “Food Radar”.

Her solution is not only topical, given the many food recalls U.S. consumers have had to grapple with this year, but it was equally impressive to watch Disha leverage Kizen’s Agentic Orchestration platform to build an AI triage system to tackle critical delays in food safety recalls - all built in under five hours. Not only does her use case solve a real world problem, but she successfully showed how a single builder can leverage data and technology to build something that truly makes life better. 

The Problem: A Dangerous 59-Day Time Gap

In the U.S., when a food firm initiates a recall, there is a critical time gap averaging 59 days before the FDA officially classifies its risk level. During this two-month window, consumers, distributors, and retailers remain unaware of potential hazards, leaving public health at risk due to manual triage processes.

The Solution: "Food Radar"

To close this critical delay, Disha built Food Radar: a workflow and end-to-end AI triage system that fast-tracks risk assessments and supply chain alerts. By unifying real-time data ingestion, multi-agent AI reasoning, and human oversight, Food Radar drastically reduces the time required to evaluate and respond to food safety threats.

Leveraging the Full Power of Kizen

What impressed the Hackathon judges most was the breadth of Kizen features Disha leveraged and integrated into a cohesive, production-ready solution in just a few hours. Key components included:

• Public Data & API Ingestion: Seamlessly pulling historical recall data directly from public FDA open-source APIs into Kizen’s system.
• Interconnected Agentic Workflows: Built 4–5 specialized agentic workflows operating together, including:
  
  • Priority Assessment Workflow: Uses four specialized agents (Historical Data, Distribution Impact, Compliance, and Orchestrator) to generate an immediate risk rating and reasoning memo. A review task will be triggered and a human reviewer will double-check the AI reasoning.
  • Complaint Early Warning Workflow: Monitors and clusters consumer complaints within a specific time period and auto-creates recall records, and triggers a priority assessment agentic workflow when volumes breach certain safety guardrails.
  • Traceability Notification Workflow: Flags high-risk recalls for individuals to review before automatically notifying affected distributors and retailers across the supply chain.
  • Review Approval Sync Workflow: Verifies whether the human reviewers approve the AI classifications.
  • Real Send Demo: Triggers live SMS notifications directly to designated human reviewers.
• LLMs & AI Reasoning: Leveraged AI actions for contextual reasoning to automatically evaluate ingested data and deduce optimal next steps.
• Objects, Records, and Custom Fields: Structured custom data models within Kizen to maintain clean, updated recall records.
• Notifications: Automated SMS and email alerts sent directly to key stakeholders and operational teams, generating follow-up tasks when human intervention is required to validate AI recommendations.
• Centralized Dashboards: Gathered incoming data and risk metrics into real-time visual dashboards for comprehensive monitoring.

The Advantage: Unification Without the Complexity

Building a system like Food Radar typically requires stitching together numerous software platforms, perhaps four or five different solutions that need to be integrated; a system of record, task management tools,  an LLM orchestration layer, notification engines, and external dashboard tools, in order to receive the comprehensive overview that Disha’s solution can provide. 

The power of Kizen lies in having every one of these capabilities baked seamlessly into a single platform. By providing a unified system of record alongside LLM integration, automated agent workflows, dashboards, and multi-channel messaging, Kizen empowered Disha to create a commercially viable, potentially life-saving application in less than five hours - and she really leveraged a strategic combination of Kizen’s agentic orchestration features available to her to build a one-of-a-kind workflow.

‍