Suspicious activity detection
Designed to help identify potentially suspicious events so teams can review them with more context.
Currently in development
AI-Assisted Security
Turn existing CCTV into AI-assisted business monitoring.
Bherundé is being built to help hotels, lodges, retail businesses, and other cash-handling environments review important events in footage they already capture — without claiming perfect detection or guaranteed prevention.
Problem
Most CCTV records events. Very little of it is actually reviewed.
Businesses already invest in cameras, especially around cash counters, lobbies, and operational areas. The footage is often used only after an incident is already known. Continuous human review does not scale, and important moments are easy to miss. Teams need a way to surface potentially relevant events so they can investigate with context.
Solution
AI-assisted monitoring on infrastructure you already own.
Bherundé is intended to analyze existing CCTV footage and help identify potentially suspicious events — including cash-handling activity and possible pocketing or theft-related behavior. The product direction includes event timestamps, incident identification, alerts, AI-generated summaries, and centralized monitoring so teams can review what matters. It is designed to detect important events for human review, not to replace security personnel.
Intended capabilities
Designed to help teams review what matters
These capabilities describe the product we are developing. They should not be read as a list of features already in production.
Cash-handling event detection
Intended to surface notable cash-counter activity for later review, not to replace cashier supervision.
Potential theft / pocketing events
Designed to help flag events that may warrant investigation. This is assistance, not a guarantee of prevention.
Timestamps, alerts, and summaries
Planned support for event timestamps, alerts, and AI-generated summaries that make review faster.
Centralized monitoring
Intended to give operators a single place to inspect flagged events across cameras.
Who it's for
First environments we are designing for
Hotels and lodges
Properties that already run CCTV and need better visibility around cash, staff areas, and guest-facing operations.
Cash-counter businesses
Operators who want AI-assisted review of cash-handling activity without ripping out existing cameras.
Retail environments
Stores that need a more practical way to review potentially important events after they occur.
Possible later environments
- Hotels
- Lodges
- Cash-counter businesses
- Retail
- Warehouses
- Manufacturing
- Hospitals
- Other commercial environments
AI and technology
Computer vision as a review layer
Bherundé is being designed around existing camera infrastructure. The technical aim is computer-vision-assisted security: detect and summarize events that may deserve human attention. Models can miss events and can also flag events that are not incidents. That is why the product is framed as AI-assisted monitoring — a tool for review, not an autonomous security guarantee.
Future vision
From cash-handling focus to broader commercial monitoring
The first product focus is hotels, lodges, retail, and other cash-handling environments. Over time, the same approach may extend to warehouses, manufacturing, hospitals, and other commercial settings. Expansion will follow evidence, not slogans.
Product status
Currently in development
Bherundé is part of the initial Manuri Innovations product portfolio and is under active development. There is no public launch date, pricing, or customer list on this site because those facts do not exist to publish yet.
Bherundé does not guarantee theft prevention or perfect detection. It is being designed to help businesses identify and review potentially important events in CCTV footage.
Talk to us about BherundéInterested in this product?
These products are being built. If the problem they address is relevant to your work, we would like to hear from you.