Looks like you have JavaScript disabled. For the full Twine experience, you will need to re-enable it.

Computer Vision Engineer with 3+ years of experience building and deploying production-scale AI/ML systems for retail inventory automation, including object detection, segmentation, and OCR pipelines. Strong end-to-end ML lifecycle ownership with PyTorch and modern MLOps practices. Proven results including achieving 97% mAP, reducing manual annotation effort by 50%, improving dataset diversity by 35%, and cutting retraining cycles by 2×. Experienced with MLflow, ONNX Runtime, Azure, Kafka-based streaming, and inference optimization to deliver reliable, monitored models in production.…Computer Vision Engineer with 3+ years of experience building and deploying production-scale AI/ML systems for retail inventory automation, including object detection, segmentation, and OCR pipelines. Strong end-to-end ML lifecycle ownership with PyTorch and modern MLOps practices. Proven results including achieving 97% mAP, reducing manual annotation effort by 50%, improving dataset diversity by 35%, and cutting retraining cycles by 2×. Experienced with MLflow, ONNX Runtime, Azure, Kafka-based streaming, and inference optimization to deliver reliable, monitored models in production.

Jaya Sravan Kala

AI Engineer, Data Engineer, Software Engineer, +2





Computer Vision Engineer with 3+ years of experience building and deploying production-scale AI/ML systems for retail inventory automation, including object detection, segmentation, and OCR pipelines. Strong end-to-end ML lifecycle ownership with PyTorch and modern MLOps practices. Proven results including achieving 97% mAP, reducing manual annotation effort by 50%, improving dataset diversity by 35%, and cutting retraining cycles by 2×. Experienced with MLflow, ONNX Runtime, Azure, Kafka-based streaming, and inference optimization to deliver reliable, monitored models in production.…Computer Vision Engineer with 3+ years of experience building and deploying production-scale AI/ML systems for retail inventory automation, including object detection, segmentation, and OCR pipelines. Strong end-to-end ML lifecycle ownership with PyTorch and modern MLOps practices. Proven results including achieving 97% mAP, reducing manual annotation effort by 50%, improving dataset diversity by 35%, and cutting retraining cycles by 2×. Experienced with MLflow, ONNX Runtime, Azure, Kafka-based streaming, and inference optimization to deliver reliable, monitored models in production.

Available to hire

Computer Vision Engineer with 3+ years of experience building and deploying production-scale AI/ML systems for retail inventory automation, including object detection, segmentation, and OCR pipelines. Strong end-to-end ML lifecycle ownership with PyTorch and modern MLOps practices.

Proven results including achieving 97% mAP, reducing manual annotation effort by 50%, improving dataset diversity by 35%, and cutting retraining cycles by 2×. Experienced with MLflow, ONNX Runtime, Azure, Kafka-based streaming, and inference optimization to deliver reliable, monitored models in production.

Skills

Experience Level

Expert

Expert

Expert

Expert

Expert

Expert

Intermediate

Intermediate

Intermediate

Intermediate

Language

Work Experience

Computer Vision Engineer at Sam's Club

February 1, 2025 - Present

Owned development of production-grade YOLO-based OBB, segmentation, and object detection models, including training/evaluation, inference optimization, and MLflow-based experiment tracking. Achieved 97% mAP across critical retail classes. Built production segmentation post-processing with VSB-guided bin splitting, NMS, and Mask IoU filtering to reduce duplicate/overlapping detections (~40%). Implemented automated image ingestion with Verint APIs plus quality validation (blur detection, fisheye filtering), increasing dataset diversity by 35% and reducing retraining frequency by 2×. Architected modular CV pipelines processing 400+ shelf cameras across deployment phases; integrated YOLOv11, DOLG embeddings, VLM-based UPC change detection, and Apache Kafka streaming for real-time inventory monitoring. Developed multi-model price-tag/UPC workflow using tiled inference and multi-pass OCR (EasyOCR → PaddleOCR → fuzzy digit normalization) with severity-tiered alerts. Productionized simila

Software Trainee at UCode Technologies LLC

September 1, 2024 - January 1, 2025

Designed and developed ETL pipelines to ingest and transform structured data from SQL and AWS S3 into clean, model-ready datasets for predictive analytics. Automated data preparation workflows to reduce manual preprocessing and accelerate experimentation. Developed and evaluated Random Forest classification models, tuning precision-recall tradeoffs to match business performance objectives. Implemented automated data validation workflows to detect missing/inconsistent/anomalous records and improve overall training data quality.

Program Analyst Trainee at Cognizant

February 1, 2021 - July 1, 2022

Managed and monitored enterprise ETL workflows across MySQL and Snowflake, ensuring reliable data availability and SLA compliance. Developed SQL transformations and Bash automation scripts using Delta Lake to improve data quality and reduce downstream dashboard processing time. Contributed to Snowflake schema optimization and data warehouse enhancements to improve reporting efficiency and analytical query performance.