Sunkara Bhavana
Machine Learning Engineer | Generative AI & NLP | MLOps | Computer Vision | AWS | PyTorch | Delivering Scalable AI Solutions
- Role
- Ml Software Engineer at 摩根大通
- Location
- Hyderabad, TG, IN
- LinkedIn followers
- 500 followers
About Sunkara Bhavana
Machine Learning Engineer with 1.7 years of experience developing and deploying ML-based…
Experience
Ml Software Engineer
Aug 2024 — Present
Designed and deployed scalable machine learning models for financial transaction monitoring and fraud detection using Python, Scikit-learn, and PyTorch, reducing false positives by 23%. Built and optimized NLP pipelines leveraging LLMs (GPT, BERT) via Hugging Face and SpaCy to classify financial documents and extract key entities with over 92% accuracy. Developed a distributed training framework using Apache Spark and MLflow, accelerating training speed by 40% and supporting large-scale modeling on AWS Sagemaker and EC2. Integrated CI/CD pipelines using Jenkins, Docker, and GitHub Actions for automated testing, model deployment, and monitoring, improving model delivery time by 35%. Implemented model drift detection and performance monitoring using MLflow and custom Python scripts to ensure compliance and real-time adaptability in production. Collaborated cross-functionally with Data Engineers and Product Managers to build ETL pipelines (Kafka, AWS S3, Lambda) and support real-time model inference at scale. Developed and deployed CNN-based computer vision models for check image verification using TensorFlow and OpenCV, achieving a 94% image recognition accuracy. Participated in bi-weekly Agile sprints and code reviews, ensuring quality and maintainability of ML software across the team. Developed and deployed real-time fraud detection models using PyTorch and Scikit-learn on AWS SageMaker, improving fraud detection accuracy by 23%. Built and integrated NLP pipelines using Hugging Face Transformers and BERT for automated document classification, increasing processing efficiency by 40%. Designed containerized ML workflows with Docker and Kubernetes, enabling scalable and reproducible model deployment across multiple banking systems.
Education
University of North Texas
Master, Data Science
GITAM Deemed University
Bachelor of Technology - BTech, Computer Science
2018 — 2022
Find verified contacts for anyone on LinkedIn
Unifers gives sales teams verified emails and direct dials, enriched profiles, and outreach that lands in the inbox.
Free plan included · No credit card required
This profile is compiled from publicly available professional sources. Unifers is not affiliated with or endorsed by LinkedIn. Request removal of this profile.