Sriram Srinivasan

AI/Software Engineer pursuing MS in Computer Science at Virginia Tech

Experience

Graduate Research Assistant | Virginia Tech
📅 December 2024 - Present

Advisor: Dr. Christine Julien

AI Engineer Intern | Strategy
📅 June 2025 - Aug 2025

  • During my internship at Strategy, I built BotFlow, a full-stack AI bot workflow authoring studio that lets anyone create and manage chatbots through an easy drag-and-drop interface, no coding, no tears.

  • Built with React Flow, FastAPI, PostgreSQL, and the LangGraph execution engine, BotFlow helped cut chatbot build time by nearly 90%, turning what used to be a long weekend project into a coffee-break task.

  • I also designed BotChat, an AI co-authoring assistant powered by GPT-4o, that translates natural language instructions into executable AI bot JSON workflows, reducing manual setup effort by about 50% and raising team morale by roughly the same amount.

  • Some highlights: I got to attend Google’s AI Agentic Development Workshop in NYC and even present BotFlow to the EVP and CTO of Strategy, which thankfully went smoother than most of my first demos.

Overall, the internship helped me bridge AI engineering and software engineering, and reminded me that great tools don’t just make systems smarter, they make people’s work a little easier (and a lot more fun).

Software Engineer | Sabre Corporation
📅 Jan 2023 - Dec 2023

At Sabre Corporation, I worked on SABRE2, the company’s internal product deployment and maintenance platform that supports 700+ hosted applications and over 1,000 users worldwide. My work sat at the intersection of engineering, automation, and a fair bit of investigative debugging.

  • Modernized CI/CD pipelines for 50+ Spring Boot applications by migrating them to SABRE2, improving deployment speed and consistency across teams

  • Reduced GKE infrastructure costs by 15% by developing an automated Jenkins pipeline (using Groovy scripting) to monitor and deallocate idle Kubernetes resources and services

  • Lowered internal tooling costs by 30% by deploying an open-source knowledge-sharing solution on Sabre’s Kubernetes cluster

  • Improved system reliability by resolving 95% of SABRE2 incidents within 15 minutes, keeping SABRE2 deployments running smoothly for hundreds of engineers

  • Transitioned from Software Engineer Intern to Software Engineer, continuing to contribute to the platform engineering team at Sabre

  • Some Highlights

    • Winner, Sabre Intern Bootcamp Hackaton
    • Finalist, of Sabre’s Blitzcode Hackathon
    • Selected as one of the top 5 students from my university and among the top 50 nationwide for Sabre’s Internship Program

Working at Sabre gave me a front-row seat to large-scale DevOps systems in motion and the satisfaction of knowing that a well-tuned pipeline can make hundreds of engineers’ days a little easier.

Research Intern | Sony
📅 Oct 2021 - June 2023

Advisor: Dr. Vidhya Balasubramanian

During my undergrad, I worked on a Sony-funded Edge AI project aimed at reducing post-harvest food loss by teaching a solar polyhouse when it’s time to chill out (literally). Using Sony’s Spresense IoT board, I developed an EdgeAI-based cyber physical system to predict food dryness levels inside a solar polyhouse and automate temperature control accordingly.

  • Built and deployed AI models on the Sony Spresense board to monitor food dryness with around 75% accuracy, minimizing spoilage and improving yield
    • Leveraged Machine Learning, Deep Learning and Computer Vision techniques for on-device processing of the collected data using camera module and other sensors from the polyhouse

  • Automated the control of an exhaust fan to adjust temperature and humidity based on real-time dryness predictions, because no one likes over-dried produce.

  • Integrated a Federated Learning framework that allowed multiple edge devices to collaboratively train models without sharing local data, improving performance by about 20% while preserving privacy

  • Some Highlights
    • The only team nationwide to secure funding for four consecutive project phases.
    • Presented our work to Sony Tokyo, which was both nerve-wracking and incredibly rewarding.
    • One of just three undergraduates initially recruited to the project.

This experience taught me how much impact a small, well-placed sensor (and a lot of debugging) can have on something as essential as reducing food waste. Plus, I learned that even solar polyhouses have their mood swings.

Researcher | AMUDA Labs, Amrita Vishwa Vidyapeetham
📅 June 2022 - June 2023

Advisor: Dr. Karthi R

At AMUDA Labs, I worked on bringing Federated Learning out of the cloud and into the hands of tiny resource constrained devices. I implemented a collaborative training architecture from scratch on microcontroller units (MCUs) such as Raspberry Pi and Arduino Nano 33 BLE Sense, enabling privacy-preserving machine learning at the edge.

  • Designed and deployed a Federated Learning framework optimized for resource-constrained MCUs, reducing dependency on centralized data transfer while maintaining competitive model accuracy.

  • Published the research at ICMLDE 2023, contributing to the broader conversation on decentralized, privacy-aware AI.


Publications

Federated learning on low-power Arduino Nano33 BLE Sense to predict the length of stay using a linear regression model

Published in International Conference on Machine Learning and Data Engineering (ICMLDE 2023), 2023

  • Abstract: This study explores the use of Federated Learning (FL) on low-power Arduino Nano 33 BLE Sense microcontrollers to predict hospital length of stay while preserving data privacy. FL enables collaborative model training from decentralized data sources, with experiments analyzing the impact of data distribution, local epochs, and FL rounds on model performance. Results demonstrate that FL can be effectively implemented on resource-constrained MCUs for privacy-preserving healthcare applications.
  • Link: Publication


Projects

My Alexa

A Voice controlled Smart Lighting System using TinyML - My Alexa. Works entirely locally without internet. Imrproved privacy and latency

  • Tech Stack: Arduino C, Deep Learning, Computer Vision, TensorFlow, TF Lite, TF Lite Micro
  • Demo: Link
  • GitHub: Link

Smart Mask Enforcement System

This is a TinyML application which operates on the MultiTenant Cascading architecture. This TinyML application intelligently predicts whether you are present or not using the input it recieves through its inbuilt microphone and then predicts whether you are masked or not through the input it recieves through its camera module (the OV7670 camera module) and actuates appropriate responses.

  • Tech Stack: Arduino C, Deep Learning, Computer Vision, TensorFlow, TF Lite, TF Lite Micro
  • GitHub: Link

Boneage Prediction from Medical Images

A machine learning model for accurately predicting the bone age and gender of a person from the X-ray images

  • Tech Stack: Python, Keras, Machine Learning, Deep Learning, Computer Vision, TensorFlow
  • GitHub: Link


Education

Master of Science in Computer Science

Virginia Tech
📅 Expected Graduation: May 2026

  • Relevant Courses:
    • Fall 2024: Machine Learning, Advanced Topics in Software Engineering
    • Spring 2025: Natural Language Processing, AI Tools in Software Engineering
  • Current GPA: 3.67/4.0

Bachelor of Technology in Computer Science

Amrita Vishwa Vidyapeetham
📅 Graduated: 2023

  • Relevant Courses: Data Structures and Algorithms, Design and Analysis of Algorithms, Big Data Anaytics, Machine Learning, Deep Learning, Fundamentals of Data Science, Database Management System, Computer Networking, Computer Organization and Architecture, Software Engineering, Embedded Systems, Single Variable and Multi Variable Calculus, Discrete Mathematics, Linear Algebra, Probability and Random Processes
  • GPA: 8.92/10.0
  • Honor: Graduated in First Class with Distinction


Skills

Programming Languages

Python, Java, C, C++, Arduino C, ES6, Kotlin, Shell Scripting

Tools & Frameworks

React, Node.js, Express, HTML, CSS, JavaScript, Django, Flask, Java Spring, Java Spring Boot, Microservices

Cloud & DevOps

GCP, GKE, AWS, Docker, Kubernetes, Terraform, CI/CD, Jenkins, YAML, Groovy

AI

Numpy, Pandas, Seaborn, OpenCV, Scipy, Matplotlib, Scikit-learn, Keras, TensorFlow, PyTorch, TFLite, LLMs, VLM’s, Foundation Models

Software Management

Git, Github and Bitbucket, Unit testing, Agile and Scrum, Rally