Problem

Kirana AI addresses the operational blind spots of physical retailers, beginning with independent grocery stores. Store managers cannot continuously watch camera feeds, check shelves, investigate suspected theft, monitor safety hazards, and maintain service quality at the same time. The resulting losses include theft and other shrink, missed sales from empty shelves, workplace and customer-safety incidents, and staff time spent reviewing footage or performing manual checks. The economics are particularly painful in grocery: one industry source places average grocery profit margins at only 1.6%, while another estimates that U.S. retailers lost $90 billion to inventory shrinkage in 2025.

The clearest or “killer” use case is real-time loss prevention: detecting a customer leaving without visiting checkout, concealing merchandise, or using a known theft tactic, then alerting staff before the loss is completed. Kirana’s broader premise is that preventing one high-value loss—or identifying enough out-of-stocks to recover sales—can justify an automated monitoring layer far more cheaply than adding another employee to every store.

Product / Service

Kirana AI sells an AI store-manager platform that turns a retailer’s existing cameras into continuously monitored operational sensors, with no new cameras required. Its current website describes cloud-based processing that analyzes video in real time and sends alerts to staff. The flagship Theft Detection Platform covers no-scan exit detection, concealment and suspicious-behavior alerts, out-of-stock monitoring, health and safety compliance, and an analytics dashboard. The company claims easy setup and ROI in 30 days, although that ROI claim is not independently substantiated in the public material reviewed.

The platform is designed to extend beyond loss prevention. Kirana offers custom “AI employees” for workflows such as lobby management, queue and service monitoring, QSR kitchen and drive-through operations, and task verification. Its longer-term full-stack vision includes integrating POS and ordering data so the AI manager can help automate pricing research, ordering, assortment decisions, and delegation. The benefit is a software-driven layer of store oversight that can respond immediately to events while reducing manual monitoring and helping managers protect revenue and focus on higher-value customer and operating decisions.

Market

Kirana AI competes in the overlap between retail loss-prevention software, AI video analytics, computer vision, and store-operations automation. Its initial beachhead is grocery, but its stated use cases also cover convenience stores, broader retail, and quick-service restaurants. Comparable solutions include Spot AI and Everseen in AI-enabled retail video and loss prevention, Lumana in AI video security for grocery, and established camera and analytics vendors such as Axis. Kirana’s differentiation is its attempt to package loss prevention, inventory, safety, service, and eventually POS-driven management into a single “AI store manager,” rather than offering only a security or camera-analytics product.

The company appears to be an early-stage, venture-backed startup rather than a scaled commercial vendor. Y Combinator lists it as founded in 2025, active, based in New York, and part of its Spring 2025 batch, with a two-person founding team. PitchBook reports $500,000 raised from Y Combinator and Sand Hill North; its financing table labels the 2025 entries “Generating Revenue,” but the same profile leaves current revenue blank and the reviewed public sources disclose no customer count, named deployments, or revenue figure. The best-supported conclusion is therefore limited public traction with some indication of early commercial activity, not a confidently established revenue scale—and not enough evidence to call the company definitively pre-revenue.