Skip to content
View Hemasri20's full-sized avatar

Block or report Hemasri20

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don't include any personal information such as legal names or email addresses. Markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
Hemasri20/README.md

Hi, I'm Hema Sri Perumalla 👋

AI/ML Engineer | LLMs | MLOps | Cloud (AWS/GCP)

I build machine learning systems, deploy production-ready pipelines, and develop LLM/RAG-based intelligent applications. My experience spans fraud detection, healthcare predictive analytics, and enterprise-grade AI solutions.


🚀 About Me

  • AI/ML Engineer with 3 years of hands-on experience in banking & healthcare
  • Skilled in building end-to-end ML pipelines, deploying models to AWS, and tuning deep learning/LLM systems
  • Experienced with PyTorch, TensorFlow, Scikit‑learn, Hugging Face, LangChain, and RAG architectures
  • Passionate about MLOps, LLMs, and scalable cloud architectures
  • Currently completing MS in Computer Science (May 2025)

🔥 Featured Projects

📌 1. Fraud Detection ML Pipeline (Banking)

End-to-end pipeline using PyTorch/Sklearn + AWS SageMaker. Includes:

  • Feature engineering, ETL (Airflow/Glue), Snowflake integration
  • Model retraining + MLflow versioning
  • Drift monitoring + Power BI dashboards

🔗 Repo Coming Soon


📌 2. Healthcare Patient Risk Prediction (DL + NLP)

TensorFlow/Keras deep-learning model for predicting patient risk.

  • PySpark preprocessing of 20M+ EHR records
  • BERT-based clinical NLP pipeline
  • Flask API + ECS deployment + CloudWatch monitoring

🔗 Repo Coming Soon


📌 3. LLM Fine-Tuning + RAG Chatbot

Domain‑specific chatbot using:

  • HuggingFace Transformers
  • FAISS vector search
  • LangChain agents
  • Streamlit front‑end

🔗 Repo Coming Soon


🛠️ Tech Stack

Languages: Python, SQL, R, Java, Scala
ML/DL: PyTorch, TensorFlow, Keras, Scikit‑learn, XGBoost
NLP/LLM: Transformers, BERT, GPT, LangChain, RAG
Data Engineering: Pandas, NumPy, PySpark, Airflow, Kafka
MLOps: MLflow, Docker, Kubernetes, SageMaker, GitHub Actions, Jenkins
Cloud: AWS (S3, Lambda, ECS, Glue, SageMaker), GCP
Databases: Snowflake, PostgreSQL, MongoDB, Redshift
Visualization: Power BI, Tableau, Matplotlib


📬 Connect With Me


Thanks for visiting my GitHub! Feel free to explore my projects or reach out for collaboration.

Pinned Loading

  1. fraud-detection-pipeline fraud-detection-pipeline Public

    A demo machine learning pipeline for transaction fraud detection using Pandas, Scikit-learn, and feature engineering.

    Python

  2. LLM-CHATBOT LLM-CHATBOT Public

  3. patient-risk-prediction patient-risk-prediction Public

    A demo deep learning model using TensorFlow/Keras to predict patient health risk using synthetic EHR data.

    Python