Freelance Project Posting – Local AI Integration on Raspberry Pi 5
Budget: $750 – $1,500 USD
We are looking for a developer with proven experience in computer vision and edge-device deployment to build an offline-first solution for a Raspberry Pi 5 (Raspberry Pi OS 64-bit) that can:
Detect and classify passing vehicles via a USB camera:
car, pickup, truck, bus, motorcycle, courier/parcel van, ambulance, police, fire engine, taxi, or private vehicle.
Recognize license plates and log them to an external SQL database with the following fields: date, time, vehicle type, color, and plate number.
Operate in dual modes
Offline: process and store data locally.
Online: sync in real time with the remote SQL database and publish each event to an external MQTT EMQX broker.
Deliverables – choose one of the following:
Docker image ready for local download and installation.
Bootable microSD image for the Pi 5.
Scope & Technical Requirements
Compact AI model that runs fully offline (TensorRT, OpenVINO, or other optimized framework preferred).
Well-documented code (Python preferred) with automated deployment scripts.
Secure configuration for SQL and MQTT connections (environment variables).
Installation and user guide.
Timeline: 7 calendar days after award, with final verification on a physical device.
Interested? Please include in your proposal:
A concise work plan.
Technologies and models you intend to use.
Portfolio or links to similar projects.
Estimated budget and availability. A visual demo of something similar, would be much appreciated.
Feel free to ask for any additional information (test credentials, database or broker details). Let’s collaborate and make this project a success!
PD: The freelancer must also be willing, upon delivery, to sign an intellectual-property assignment agreement and a non-disclosure agreement (NDA).
Detect and classify passing vehicles via a USB camera:
car, pickup, truck, bus, motorcycle, courier/parcel van, ambulance, police, fire engine, taxi, or private vehicle.
Recognize license plates and log them to an external SQL database with the following fields: date, time, vehicle type, color, and plate number.
Operate in dual modes
Offline: process and store data locally.
Online: sync in real time with the remote SQL database and publish each event to an external MQTT EMQX broker.
Deliverables – choose one of the following:
Docker image ready for local download and installation.
Bootable microSD image for the Pi 5.
Scope & Technical Requirements
Compact AI model that runs fully offline (TensorRT, OpenVINO, or other optimized framework preferred).
Well-documented code (Python preferred) with automated deployment scripts.
Secure configuration for SQL and MQTT connections (environment variables).
Installation and user guide.
Timeline: 7 calendar days after award, with final verification on a physical device.
Interested? Please include in your proposal:
A concise work plan.
Technologies and models you intend to use.
Portfolio or links to similar projects.
Estimated budget and availability. A visual demo of something similar, would be much appreciated.
Feel free to ask for any additional information (test credentials, database or broker details). Let’s collaborate and make this project a success!
PD: The freelancer must also be willing, upon delivery, to sign an intellectual-property assignment agreement and a non-disclosure agreement (NDA).