Poojitha Y.
AI/ML Engineer | ML/DL Expert (Transformers, CNNs, RNNs) | LLMs | NLP | AWS • Azure • Kafka • TensorFlow • PyTorch | MLOps | Scalable Recommender Systems
- Role
- Ai Engineer at Wells Fargo
- Location
- New York, NY, US
- LinkedIn followers
- 500 followers
About Poojitha Y.
Highly experienced AI Engineer with 3+ years of expertise in Data and Machine Learning, specializing in developing and deploying advanced models using Python, TensorFlow, Keras, and scikit-learn. Proficient in data pre-processing, feature engineering, and model evaluation. Skilled in data visualization with Tableau and Power BI, and experienced in handling large datasets and implementing data pipelines. Proven ability to integrate AI solutions with AWS and Azure.
Experience
Ai Engineer
Jan 2024 — Present · NY, US
Developed and deployed a cutting-edge recommendation system leveraging advanced machine learning and deep learning algorithms (RNN, CNN, Transformers) in TensorFlow and Keras, resulting in a 35% boost in recommendation accuracy and a 25% enhancement in recommendation relevance.• Utilized Python (Pandas, NumPy, Scikit-learn) to preprocess, engineer, and tune models, optimizing prediction accuracy and performance on over 1TB of data through techniques such as data augmentation and dimensionality reduction.• Architected and managed scalable data warehousing solutions by integrating AWS services (Amazon S3, Redshift, Glue), resulting in a 45% reduction in data retrieval times and a 20% increase in data processing efficiency.• Leveraged Apache Kafka for real-time data streaming, handling over 1 million events per day, and built robust ETL pipelines that enhanced data integration efficiency by 60• Applied advanced NLP techniques (BERT, Word2Vec) and reinforcement learning algorithms integrated with human feedback to boost recommendation relevance by 30% and increase user engagement by 40%.• Developed and deployed cutting-edge deep learning algorithms for Large Language Models (LLM), Natural Language Processing (NLP), Computer Vision, and Speech, including building and scaling MLaaS platforms.• Collaborated effectively on 6 projects, accounting for 80% of the workload. Partnered with Oakridge National Lab on 2 additional projects, resulting in a 60% reduction in overall workload for the team.
Education
University of Alabama at Birmingham
Master's degree
2022 — 2024
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.