Experiences

Incedo Inc.

Software Engineer Intern

Built a Retrieval-Augmented Generation (RAG) pipeline using Ollama, ChromaDB, LangChain to power natural language Q&A over structured Root Cause Analysis data
Developed a modular CLI tool to support JSON & CSV data ingestion, enabling flexible and reproducible workflows
Enhanced user experience by adding semantic search, multi-turn context awareness, text streaminng, and an option to switch between local and cloud inference modes
Building a SQL agent to improve retrieval over numerical data and using a hybrid approach to improve quality and accuracy of generated responses

Software Engineer Intern

Gathered functional requirements with product owners, ensuring clear alignment of technical and business goals
Built a robust backend in Python & Django, exposing RESTful APIs for user authentication and data retrieval
Created frontend components, enabling users to interact with large data sets & access real-time visualizations, reducing page re-renders by 15% to increase user productivity
Containerized deployments, ensuring consistent development environments, reducing deployment failures by 20%
Streamlined testing, staging and production deployments via Gitlab CI/CD and cut release times by 30%

Teaching Assistant

Mentored 250+ students in programming, data structures & database concepts by conducting recitations and doubt sessions for assignments and projects. Improved in-class attendance by 10% by making content more engaging
Optimized, automated code grading workflows using Python scripts to reduce time-to-grade by 60%, allowing students to get quicker feedback

Software Engineer Intern

Collaborated with engineers to design and deploy a CNN-RNN model for object, activity, and posture recognition, integrating the model into a production ready web application for real-time detection
Maintained and helped develop a full-stack web application using JavaScript, Node.js, React.js and MongoDB, contributing to both backend functionality and dynamic frontend interfaces for multiple components
Used RESTful APIs to facilitate seamless communication between the frontend, backend, database, and machine learning models for real-time inference.
Optimized model performance by improving data pre-processing pipelines and tuning hyper-parameters, leading to a 10% increase in model accuracy
Setup up a CI/CD pipeline for deployments, using Git, GitLab. Used Kubernetes and AWS for scalable cloud-based infrastructure