I need a complete video-analytics setup built around Frigate (or a comparable open-source stack) that can watch two or more CCTV cameras covering the same physical area, in addition to overall area monitored by couple of CCTV cameras, and reliably label what is happening in real time. The system must: • Detect when people are present or absent, whenever someone enters or leaves, and keep short, timestamped clips for review. • Run face detection that distinguishes known from unknown faces and triggers an event. • Recognise specific actions and behaviour I care about—entry/exit and everyday activities for now—and flag them as separate event types, record time • Configure actions and create events for molecularity • Monitor nearby machinery, logging both operational status and idle periods so I can analyse utilisation later. • Provide flood and fire alerts. • Fire an n8n webhook for events Everything should be logged in a way that lets me filter and export analytics by event category (people, face match, machine status, safety incidents, etc.). I also want the architecture to stay flexible: I should be able to add new detectors or retrain models for fresh behaviours without re-engineering the whole pipeline—ideally through a clean Docker deployment. Acceptance criteria 1. All cameras stream through Frigate (or your proposed alternative) with detections visualised in its UI. 2. n8n receives distinct webhook calls for configured event type, carrying JSON with camera name, event label, and clip URL. 3. Face library is configurable via a simple folder or API. 4. A short README explains how to retrain models, add cameras, or extend n8n flows. If you’ve worked with Frigate, TensorFlow, YOLO, OpenCV, or similar stacks and can demonstrate a working PoC quickly, let’s talk. Your job is to setup, configure, document all of above and help us build, recreate the same using documentation at our end. Also, to fix any issue that is needed to meet above requirements. Acceptance Criteria: Clean repository with clear build/run instructions. Application should work flawlessly with at least 4 CCTV camera streams during your demo. Full demo video covering 100% features before delivery for review. A document covering critical test cases with its result is needed. Comprehensive document that helps us to setup everything at our end, configure things as needed. Test Scenarios: We should be able setup everything using the document, configure it ourself. Connect our CCTV feeds Define events and actions. Define or connect new models for actions. Add Faces for known faces. Event triggered for unknown faced. Event triggered for fire / flood scenarios. Events are being logged and stored in structured way, which we can access and manipulate for our need. There should be a certralized dashboard where all errors, and events are being logged. Events triggers n8n workflow, send event details in payload, which are received on n8n hooks. Deliverables Test Setup - Above test scenarios should be tested and passed on to your hosting env before delivering to me. I will need video and access to that app to verify before accepting milestone delivery. Full Source Code (GitLab) – Clean commits, meaningful messages Recommendation for suitable hardware/infra to test, help us configure the same. Technical Documentation – Installation & configuration guide, setup steps, deploy documentation, how to add/retrain models, clear instructions on how to scale to many more camera, hosting requirements for the same Demo Video – End-to-end feature demonstration Working application that work flawlessly with at least 4 CCTV streams during your demo Test Case Report – With Pass/Fail results for defined test scenarios Error-Free Build