Blog
Notes from delivery and applied AI work
Short articles covering lessons from speech, computer vision, OCR pipelines, and software implementation decisions.
Approach
Practical delivery, clear scope, and close collaboration with the team doing the work.
Focus
Custom software, integrations, product delivery, and workflow optimization.
Topics
AI
Speech
ML Ops
Computer Vision
OCR
2 articles
Apr 1, 20249 min read
Fine-Tuning Whisper Large v3 for Serbian Speech Workflows
A practical summary of adapting Whisper Large v3 for Serbian audio pipelines, focusing on dataset preparation, evaluation, and production tradeoffs.
AI
Speech
ML Ops
Sep 1, 202310 min read
YOLOv3 ROI Detection with OCR: Building a Reliable Extraction Pipeline
Designing a detection-plus-OCR pipeline is less about model choice alone and more about ROI quality, post-processing, and error handling in real documents.
AI
Computer Vision
OCR