Geeth Nischal Gottimukkala

● Designed scalable Django ORM models with AWS S3 and RDS, managing static datasets, Modbus streams, optimizer outputs. ● Built modular Django optimization platform on AWS EC2 with multi-app schemas, enabling reliable DER ops and forecasting. ● Developed Django-based data pipelines and forecasting workflows to optimize DER dispatch decisions for solar, storage, EVs, etc.

● Built automated ETL pipelines containerized with Docker, via AWS Lambda, ensuring scalable, consistent energy analytics. ● Integrated Python with Simulink on OPAL-RT via Modbus for real-time simulation, testing, and HIL validation in power systems. ● Optimized SQLite DB performance through query refactoring and indexing, accelerating simulation data processing for research. ● Designed MLOps framework (Docker/MLfl ow) standardizing model lifecycle from training to deployment, cutting release time. ● Developed forecasting models using Keras/Darts, deployed on AWS SageMaker, retrained automatically through CI/CD pipelines.

● Built an RL-based truck queue optimizer (TensorFlow) on AWS EC2/SQS via Docker, reducing wait times and improving logistics. ● Developed LSTM models for predictive maintenance on LPG plant sensors, minimizing downtime via PySpark preprocessing. ● Automated hyperparameter model tuning on AWS SageMaker GPUs, boosting prediction accuracy while reducing training time. ● Engineered a Java/WebSocket system for real-time IoT data sync to MySQL, ensuring sub-second latency for operational analytics. ● Integrated SAP ERP with automated load planning via SAP ABAP and REST APIs (FastAPI), eliminating manual dispatch errors. ● Designed an SAP-integrated emissions tracker and LPG load planner using Python and React, enabling dispatch optimization.

Online tutor for JEE advance problems/ concepts in Maths, Physics and Chemistry.

● Optimized 5G/Wi-Fi handovers by building LSTM/RNN/SoTA models (TensorFlow), reducing latency and improving signal stability. ● Designed data preprocessing pipelines using Pandas and NumPy to clean and normalize signal data, increasing model accuracy. ● Deployed ML model via Docker and Flask, enabling real-time predictions with <50ms latency in simulated network environments. ● Created a React dashboard with WebSocket integration to visualize real-time network handover analytics from Flask ML backend.

I monitor various technical societies of IIT Bhubaneswar and make sure that all the members get the opportunity to learn technical skills by getting various facilities approved by the authorities.