Hello, I'm Navisha Shetty
With a background spanning Machine Learning, DevOps, and Software Engineering, I specialize in building and scaling end-to-end AI systems. My experience ranges from large-scale data processing with PySpark and Snowflake to training and finetuning LLMs. I am proficient in deploying these solutions on cloud-native infrastructure using Docker, Kubernetes, and AWS, and I am passionate about building advanced applications like Agentic AI coding agents.
Current Projects
Current Goals
Blog Posts
Thoughts on development, technology, and life
Resume
Professional Summary
Proven expertise building and scaling production AI systems and cloud infrastructure with 3+ years of hands-on experience across data science, MLOps, and software engineering.
Work Experience
Software Engineer Intern
Jan 2025 - Aug 2025LG Energy Solution Vertech, Westborough, MA
- Designed coding agent using ReAct framework for real-time Day-Ahead Market energy data queries with Phoenix tracing and chat UI
- Built scalable Snowflake data pipeline using Snowpark API in Python with automated AWS Lambda execution
- Enhanced FASTAPI + Streamlit Energy Market Optimization tool with live feeds, AWS S3 configurations, and JWT authentication
- Applied predictive analytics on historical battery usage data to identify energy discharge cycle patterns
Software Engineer
Oct 2021 - Aug 2023TEKSystems Global Services, Bangalore, India
- Provisioned GPU instances on AWS EC2 within Kubernetes clusters for accelerated AI/ML processing
- Designed Apache Airflow DAGs to automate data workflows, improving operational efficiency
- Developed PySpark scripts for ML project migration to Spark platform, achieving 25% speed improvement and 15% resource reduction
- Deployed MLFlow on Kubernetes as managed service for 50+ ML models with Terraform-based AWS Load Balancers
Associate Data Science Engineer
Jan 2020 - Oct 2021Pitney Bowes, Pune, India
- Implemented NLP solutions (Word2Vec, GloVe) for automated HS10 Classification tax and shipping calculations
- Trained ANN and RNN models using PyTorch, enhancing parcel weight estimation accuracy by 25%
- Deployed Estimated Delivery Date model using CatBoost algorithm, achieving 98% F1 score performance
- Eliminated manual intervention in commodity classification through intelligent automation
Education
Master of Science in Data Analytics Engineering
Dec 2025Northeastern University, Boston, MA
Coursework: Probability and Statistics, MLOps, Prompt Engineering and AI, Data Mining, Computation and Visualization, Database Management
Bachelor of Technology in Information Technology
May 2020Manipal University Jaipur, India
Coursework: Data Structures, Data Science, Data Mining and Warehousing, Advanced Machine Learning
Technical Skills
Programming & Frameworks
Python
PySpark
SQL
TensorFlow
PyTorch
PostgreSQL
Spark
DevOps & MLOps
AWS
Docker
Kubernetes
Terraform
Apache Airflow
Git
CI/CD
Linux
MLflow
AI/ML & Data
ReAct Agents
Machine Learning
LangChain
LLMs
Agentic AI
NLP
Deep Learning
Certifications
Snowflake Snowpro Core
DeepLearning.ai Agents
Publications
Machine Learning for Prognosis of Life Expectancy and Diseases
Research paper published in the IJITEE Journal in August 2019 edition on applying machine learning techniques for healthcare prediction models