Available to hire
I am Akhil Chaitanya Ghanta, an AI/ML engineer with 3+ years of experience developing and deploying machine learning and deep learning models for NLP, computer vision, and generative AI. I am proficient with TensorFlow, PyTorch, Keras, and Hugging Face Transformers, and I excel in feature engineering, model optimization, and transfer learning.\n\nI build scalable data pipelines using Spark, Hadoop, and Kafka; deploy models with Docker, Kubernetes, and FastAPI; work with MLOps tools like MLflow, Airflow, and Kubeflow; and leverage cloud ML services on AWS, Azure, and GCP. I have strong SQL/NoSQL skills and create actionable insights and dashboards with Tableau and Power BI.
Skills
Language
English
Fluent
Work Experience
AI/ML Engineer at JP Morgan Chase
January 1, 2025 - PresentBuilt distributed fraud-scoring with Apache Spark MLlib on Hadoop to enable real-time scoring on 100M+ daily records. Optimized SQL queries across Oracle/MySQL to retrieve and preprocess high-volume transactions, with data quality checks and lineage tracking. Trained autoencoders for anomaly detection, achieving a 30% improvement in fraud catch rates at a fixed precision. Applied TextBlob and Hugging Face transformers (BERT, RoBERTa) for sentiment/intent analysis on unstructured text, improving customer-behavior prediction accuracy by 22% with calibrated thresholds and monitoring. Deployed YOLO computer-vision models to detect document forgery in KYC scans, reducing compliance risk by 18%. Delivered sprint-scoped model enhancements under Agile, and orchestrated ML pipelines with Airflow, Vertex AI, and MLflow for experiment tracking and automated retraining. Created fraud-risk dashboards in Tableau and visualized model performance with Matplotlib to accelerate decision-making for the c
AI/ML Engineer at Mindtree
July 1, 2023 - October 15, 2025Developed end-to-end ML pipelines in Python leveraging NumPy, Pandas, and PostgreSQL for data preprocessing, feature engineering, and model deployment, reducing manual data preparation time by 35%. Built supervised and unsupervised models for customer segmentation and fraud detection, improving prediction accuracy by 28%. Planned LSTM-based deep learning models for serial transaction pattern analysis, achieving a 25% boost in anomaly detection precision. Implemented GANs for synthetic document data generation, enhancing training datasets and improving NLP model robustness by 20%. Applied transfer learning with PyTorch on pre-trained embeddings, reducing model training time by 40% while maintaining accuracy. Leveraged NLTK for preprocessing and VADER sentiment, with spaCy for NER and scikit-learn for topic modeling to derive document-level insights. Integrated OpenCV-based CV modules for document image classification and OCR, improving data extraction accuracy. Built real-time Kafka pip
AI/ML Engineer at JP Morgan Chase
January 1, 2025 - November 18, 2025Built distributed fraud-scoring with Apache Spark MLlib on Hadoop, enabling real-time scoring on 100M+ daily records. Optimized SQL queries across Oracle/MySQL to retrieve and preprocess high-volume transactions, reduced query runtime and unblocked downstream training and scoring with data quality checks and lineage tracking. Trained autoencoders for anomaly detection, achieving 30% improvement in fraud catch rates at fixed precision. Applied TextBlob and Hugging Face transformers (BERT, RoBERTa) for sentiment/intent on unstructured text, improving customer-behavior prediction accuracy by 22% with calibrated thresholds and monitoring. Deployed YOLO CV models to detect document forgery in KYC scans which reduced the compliance risk by 18%. Delivered sprint-scoped model enhancements under Agile, meeting 95% on-time completion. Orchestrated ML pipelines with Airflow, Vertex AI, & MLflow, streamlined experiment tracking and automated model retraining. Created interactive fraud-risk dashboa
AI/ML Engineer at Mindtree
July 1, 2023 - July 1, 2023Developed end-to-end ML pipelines in Python leveraging NumPy and PostgreSQL for data preprocessing, feature engineering, and model deployment, reducing manual data preparation time by 35%. Built supervised and unsupervised learning models for customer segmentation and fraud detection, improving prediction accuracy by 28%. Planned LSTM-based deep learning models for serial transaction pattern analysis, achieving a 25% boost in anomaly detection precision. Implemented GANs for synthetic document data generation, enhancing training datasets and improving NLP model robustness by 20%. Applied transfer learning with PyTorch on pre-trained embeddings, reducing model training time by 40% while maintaining high accuracy. Leveraged NLTK for NLP preprocessing and VADER sentiment, with spaCy for NER and scikit-learn for topic modeling, producing document-level insights and reducing manual review. Integrated OpenCV-based CV modules for document image classification and OCR. Built real-time Kafka pi
Education
Master of Science in Computer Science in AI at University at Buffalo (SUNY Buffalo)
January 11, 2030 - January 1, 2025Bachelor of Science in Computer Science at Vellore Institute of Technology (VIT), Vellore, India
January 11, 2030 - May 1, 2023Master of Science in Computer Science in AI at University at Buffalo (SUNY Buffalo)
January 1, 2025 - November 18, 2025Bachelor of Science in Computer Science at Vellore Institute of Technology (VIT)
January 11, 2030 - May 1, 2023Qualifications
Industry Experience
Software & Internet, Financial Services, Professional Services, Media & Entertainment
Skills
Hire a AI Engineer
We have the best ai engineer experts on Twine. Hire a ai engineer in Dallas today.