Case Studies
A selection of projects across enterprise knowledge systems, document intelligence, computer vision, and medical AI.
These case studies document engineering work by Relytic’s founder and collaborators, including work completed before Relytic was established. They demonstrate the engineering capabilities and reliability practices now brought together under Relytic.
The common thread is not a particular model or framework. It is an engineering approach built around measurable performance, failure analysis, and production use.
01
Fortune Global 500 industrial company / Enterprise RAG / Document Intelligence / AI Evaluation
A production enterprise knowledge system serving hundreds of employees across hundreds of thousands of documents in seven languages — with thousands of queries every day.
End-to-end product pipeline
Read case study02
Fortune Global 500 industrial company / Computer Vision / Production ML
An industrial satellite/aerial imagery detection system improved through model benchmarking, structured error analysis, dataset corrections, targeted augmentation, and production optimization.
84.5% → 92.0%mAP@50
Read case study03
Mknoon AI / Medical AI / Computer Vision / Cloud Deployment
End-to-end medical-imaging AI work spanning dataset preparation, model development, evaluation, backend APIs, Google Cloud deployment, and product integration.
95.5% F1chest X-ray classification
Read case study04
Golf Vision AI / Video Analysis / AI Feedback / Cloud Deployment
Golf Vision AI turns a golfer’s uploaded swing video into structured analysis and personalized feedback through a complete production backend.
End-to-end product pipeline
Read case studyNext step
The architecture changes with the problem; Relytic’s focus on measurable reliability does not. Book a 30-minute conversation to discuss what you are trying to build or improve.