Founding Senior Computer Engineer

About the Role
This position focuses on AI match analysis for youth football, leveraging fixed cameras at the pitch to record every session. The models detect key events like passes, shots, and goals, track players, and read jersey numbers from low-resolution footage. This technology is currently operational at a pilot academy in Dubai, where it generates match statistics, highlights, and detailed per-player records. The role involves enhancing the live sports models and developing an additional product in a second industry using the same technical foundations.

What You Own

• The complete computer vision (CV) platform, including camera ingestion, detection and tracking, optical character recognition (OCR), classification, and reporting pipelines. This platform supports an existing product and a new one set for launch.
• From the first week, you will be shipping updates to a live product, improving the accuracy of the sports models with real users, while constructing the deployment for the second industry on the same stack.
• Responsibilities include architecture design, model selection and training strategy, annotation and dataset operations, evaluation methodology, and managing the on-prem/cloud distribution.
• Oversee the initial deployments for the new industry and enhance both products as a unified platform.
• Provide technical leadership to the engineering team that will be built under your guidance.

Hiring Process

• Initial conversation: Present a CV system you developed, including your decisions, challenges, and metrics.
• In-depth technical discussion with senior engineers.
• Case study involving real footage: specification and prototyping of a detection task.
• Final conversation with the founders, followed by an offer.
• The entire process is expected to conclude within two weeks for the right candidate.

What We Need to See

• A minimum of 5 years of experience in building computer vision systems deployed in production, specifically those used by real users rather than theoretical notebooks.
• Strong proficiency in Python and PyTorch, with hands-on experience in modern detection and tracking techniques (e.g., YOLO family, ByteTrack/DeepSORT).
• Experience with video pipelines, including RTSP/IP cameras, scheduled or streaming inference, and fluency with tools like ffmpeg.
• Proven background in building datasets, managing annotation operations, and accurately measuring model performance against ground truth.
• Experience in fine-tuning open-source models (in the realm of detection and OCR, specifically PaddleOCR or TrOCR) and advancing to developing proprietary models based on validated annotations, including a functioning annotation-review loop.
• Familiarity with the modern computer vision tooling ecosystem, such as Roboflow or equivalent for dataset and annotation operations, experiment tracking, and model registries.
• A sense of ownership that drives you to transform ambiguous operational goals into fully functional systems without waiting for detailed specifications.

Strong Pluses

• Experience with OCR, action recognition, or pose estimation in a production context.
• Background in sports, retail, industrial, or facility camera analytics.
• Knowledge of edge deployment (e.g., Jetson or similar) and cost-effective inference engineering.
• A solid understanding of camera and IoT technology, including sensor and lens selection, PoE networking, ONVIF/RTSP, NVRs, and weatherproof housing options. You should be capable of specifying a site as well as model requirements.

Contract and Benefits

• This is a full-time role, with a remote-friendly approach—hiring the best talent regardless of geographical proximity.
• Working hours will need to overlap with the Gulf region, and travel to Dubai will be necessary for installations and milestone weeks.
• Competitive compensation package, including options.