This project is described in more detail below under my job experiences. I’m mentioning it here separately because I spent almost 2 years working on the system and developed deep expertise in self-learning AI architectures, large-scale data pipelines, and SAP and AWS-based integrations. Unfortunately, due to strict security and confidentiality agreements, I’m not able to showcase the actual project or share visual examples. The images shown here are AI-generated and not from the real system.
When I joined the project, a conceptual design for a self-learning facial recognition system was already in place, and I was expected to follow it. However, after carefully reviewing the architecture and performance bottlenecks, I convinced both my manager and project lead to give me two additional weeks to redesign the concept.
That decision paid off:
✅ 𝗦𝘆𝘀𝘁𝗲𝗺 𝗲𝗳𝗳𝗶𝗰𝗶𝗲𝗻𝗰𝘆 𝗶𝗻𝗰𝗿𝗲𝗮𝘀𝗲𝗱 𝗯𝘆 𝗮 𝗳𝗮𝗰𝘁𝗼𝗿 𝗼𝗳 𝟮𝟬𝟬𝟬
✅ 𝗥𝗲𝗰𝗼𝗴𝗻𝗶𝘁𝗶𝗼𝗻 𝗮𝗰𝗰𝘂𝗿𝗮𝗰𝘆 𝗶𝗺𝗽𝗿𝗼𝘃𝗲𝗱 𝗯𝘆 𝟯𝟲%
The system became the backbone of a large-scale image management application used by a public organization maintaining over 400,000 photographs of historical events involving Federal Chancellors and Presidents. It operates on a modern SAP architecture (HANA, HANA XSA, Data Intelligence) and includes features like automated image import, captioning, tagging, and order management.
My role included:
- Designing the self-learning facial recognition system on SAP Data Intelligence.
- Conducting a feasibility study to ensure technical viability and scalability.
- Selecting, customizing, and integrating AI models (TensorFlow, PyTorch) for facial recognition and image tagging.
- Developing efficient Python data pipelines for real-time data flow and CRUD operations.
- Building and connecting REST APIs and Docker-based environments to streamline deployment.