Computer Vision / Machine Learning Engineer

We are seeking a highly experienced Computer Vision / Machine Learning Engineer to develop a human intrusion detection system from scratch as part of a governmental project.

This is a structured, security-focused engagement. The selected engineer will be responsible for designing and implementing a proprietary intrusion detection module that will serve as the foundation for a larger AI video analytics platform.

Successful delivery of this IDS module will lead to expanded development of additional models, long-term collaboration, increased financial scope, and potential relocation opportunities depending on performance and project progression.

Project Context

This engagement is related to a governmental request involving AI-based security analytics.

The initial objective is to develop a high-performance Intrusion Detection System (IDS) capable of real-time human detection within defined zones and virtual perimeters.

If Phase 1 (Intrusion Detection) is delivered successfully and validated, the roadmap includes development of additional models and modules within the same system architecture.

Core Requirements

The engineer must:

Design and implement the intrusion detection architecture from scratch
Structure the full training pipeline
Implement polygon-based restricted zone logic
Implement virtual line-crossing logic
Support optional direction-based rules
Handle partial body visibility cases
Exclude animals and environmental motion
Generate structured JSON event outputs
Ensure real-time or near real-time inference

This is not a plug-and-play integration project. We expect clean architectural thinking and custom logic development.

Use of open-source detection backbones (YOLO, DETR, etc.) is acceptable if technically justified, but the intrusion logic, event system, and pipeline must be custom-developed.

Technical Expectations

Strong background in Python
Advanced experience with PyTorch or TensorFlow
Experience in real-time video processing (RTSP)
Experience in detection + tracking systems
Experience optimizing inference latency
Ability to build modular, production-ready code
Experience with Docker/containerized deployment preferred

Infrastructure & IP

All development, training, datasets, model weights, and repositories will be managed within a controlled infrastructure environment due to the nature of the project.

Deliverables (Phase 1 – IDS)

Full source code
Training and inference scripts
Trained model weights
Performance evaluation report (Precision, Recall, F1)
Deployment documentation

Performance targets will be defined and agreed upon before final validation.

LONG TERM OPPORTUNITY FOR THE REST
This is not a short-term isolated task.

If the Intrusion Detection module is successfully delivered and validated:

Additional AI models will be developed
Scope and financial compensation will significantly increase
Long-term structured collaboration is expected
Potential relocation options may be discussed depending on project expansion

We are looking for someone who sees this as a strategic opportunity, not a one-off freelance task.


Please include:

Relevant computer vision projects
Your proposed architectural approach
Estimated timeline (preferably 3-4 weeks)

BUDGETING; TBD

telegram: контакт скрыт