Skip to content
Résumé
View résumé

Open to AIML & software engineering roles

Keerthirajan Senthilkumar

AIML + Software Engineering

I build AI end to end, from fine-tuning and evaluating models to the agents, retrieval and full-stack products around them.

  • Apple AIML Intern
  • MS CS · GWU 2026
  • San Francisco, CA
Explore my work View résumé

About

AIML / Software Engineer

Keerthirajan Senthilkumar

I am a recent Master's in Computer Science graduate from The George Washington University, specializing in AIML and software engineering. My work sits at the intersection of machine learning systems, evaluation design, and practical full-stack development, with recent industry experience as an AIML Machine Learning Engineer intern on Apple's Siri Core Modeling team.

I build data-driven products with a focus on reliable model behavior, clean engineering, and measurable user value. My recent work includes LLM evaluation workflows, transformer-based modeling, mobile applications, applied ML projects, and responsive web experiences.

  • SEAS Dean's Award
  • Apple AIML Internship
  • AIML + Full Stack Projects
San Francisco, CAOpen to relocation

Education

One throughline: keep learning.
GWAug 2024 - May 2026

Graduate

Master's in Computer Science, AIML & SWE Specialization

The George Washington University, USA

  • SEAS Dean's Award / Dean's Merit Scholarship50% tuition scholarship, totaling over $38,000.
  • Relevant Coursework: Design and Analysis of Algorithms, Advanced Software Paradigms, Computer System Architecture.
SVCE, Sriperumbudur emblemAug 2019 - May 2023

Undergraduate

Bachelor's in Computer Science & Engineering

Sri Venkateswara College of Engineering, India

  • Completed 3 internships alongside my undergraduate studies.
  • Volleyball captain and Youth Red Cross volunteer.
  • Relevant Coursework: Data Structures, Natural Language Processing, Artificial Intelligence, Machine Learning and Algorithms, Business Intelligence, Advanced User-interface and Technology.

Experience

Where I've worked

From web development to modeling and evaluation work on Siri.

  1. AIML Machine Learning Engineer Intern

    AppleSeattle, WA

    May 2025 - Aug 2025
    • Worked on Siri Core Modeling initiatives involving transformer-based modeling, model fine-tuning, evaluation workflows, and performance benchmarking.
    • Built and evaluated AIML pipelines using PyTorch, Hugging Face Transformers, LoRA/PEFT, Sentence Transformers, and vector embeddings.
    • Designed deterministic LLM programming and LLM-as-judge evaluation workflows to support reproducible research and model quality analysis.

  2. Machine Learning Intern

    TCSChennai, India

    Nov 2021 - Dec 2021
    • Engineered a web app for real-time Twitter sentiment analysis, scraping product data to provide actionable insights for critical decision-making. Reduced inference runtime by 40% by incorporating data techniques such as subjectivity and polarity.
    • Leveraged Tweepy API for real-time Twitter data integration. Improved model prediction accuracy by 18% using TextBlob over traditional Machine Learning models.

  3. Web Development Intern

    The Sparks FoundationSingapore - Remote

    May 2021 - Jun 2021
    • Devised a fully responsive charity donation website integrating payment gateway with seamless transaction workflows and automated email receipts. Recognized for achieving 98% unit test code coverage for the core logic.
    • Utilized modern JavaScript frameworks and Bootstraps to enhance user experience and mobile compatibility.
    • Mentored a team of 5+ developers, facilitating pair programming sessions to improve code quality and foster collaboration.

Work

Selected work

Machine learning, generative AI, and the full-stack software that ships them. Open a project for the full case study.

11 projects

PitchPulse previewCase study
Hackathon / Sports AI

PitchPulse

AI sports science assistant for player readiness, injury risk, rPPG vitals, movement analysis, and tactical XI recommendations.

  • Flutter
  • FastAPI
  • Gemini
  • rPPG
View on GitHub
Mobile + Firebase

MoodBites

Kotlin Android app that recommends and tracks meals based on mood signals and user history.

  • Kotlin
  • Firebase
  • Android
View on GitHub
Behind the project

MoodBites

Team Lead · Sep 2024 - Nov 2024

  • Led a team of 5 to develop "MoodBites," a Kotlin-based Android application that recommends and tracks user meals based on their moods, enhancing mental well-being through data-driven insights.
  • Designed and implemented a scalable database with Firebase for efficient storage and real-time management of user meals, mood logs, and order information, including automated email receipts for order confirmations.
  • Optimized app performance, reducing load times and enhancing responsiveness to ensure a seamless user experience across devices.
Medical Imaging

GI Tract Segmentation

Deep learning project for segmenting gastro-intestinal tract organs and tumor regions from MRI scans.

  • Deep Learning
  • Computer Vision
  • Medical AI
Behind the project

GI Tract Segmentation

Team Lead · Nov 2022 - Mar 2023

  • Led a team of 3 developers to build ML model for segmenting Gastro-Intestinal Tract organs from tumors through MRI scans.
  • Applied Deep Context Metric Learning for advanced image clustering and processing—improved survival rates by 8% through targeted treatment focused on tumors rather than surrounding organs.
  • Optimized backend scalability to handle 38,000 masked 16-bit image datasets, reducing processing time from 5 hours to 2 hours for smoother operation.
Applied ML

Efficient Water Quality Analysis

End-to-end model for water quality prediction using turbidity, hardness, and related environmental factors.

  • XGBoost
  • IBM Cloud
  • ML Pipelines
View on GitHub
Behind the project

Efficient Water Quality Analysis

Team Lead · Apr 2022 - Oct 2022

  • Designed an end-to-end model to analyze water quality using multiple factors such as turbidity and hardness.
  • Incorporated XGBoost to optimize memory usage and processing speed, resulting in 20% increased model's prediction accuracy.
  • Reduced computational load by using hyperparameter tuning and adjusting maximum depth and minimum sample split.
  • Deployed the model on IBM Cloud to optimize scalability by 1.5x better than local run.
Internship Project

Twitter Sentiment Analysis

Real-time sentiment analysis web app for extracting product insights from Twitter data.

  • Tweepy
  • TextBlob
  • Python
  • Flask
Behind the project

Twitter Sentiment Analysis

The purpose of this project is to perform real-time sentiment analysis on Twitter data. Users can input either a Twitter username or a hashtag, and the application fetches relevant tweets using the Twitter API. By analyzing these tweets with TextBlob, it categorizes them as Positive, Negative, or Neutral and calculates the percentage for each sentiment. The insights can help businesses make data-driven decisions by understanding public sentiment.

Python, Tweepy, Textblob, Flask

Due to confidentiality agreements with TCS, I cannot share specific details about the projects I worked on during my internship. However, I can provide the tech stacks that I used and the project description below.

Web Development

Charity Donation Website

Responsive donation website with payment gateway integration and automated email receipts.

  • JavaScript
  • Bootstrap
  • Payments
View on GitHub

Skills

Tools I work with

  • PyTorch (used in Apple, Siri Core Modeling, Historical Event Narrator, Face Cloaking)
  • TensorFlow
  • Scikit-Learn (used in Coding Problem Recommender)
  • Sentence Transformers (used in Apple, Coding Problem Recommender)
  • Neural Networks
  • Data Mining
  • Model Evaluation
  • Hugging Face (used in Apple, Historical Event Narrator)
  • LoRA / PEFT (used in Apple, Historical Event Narrator)
  • Mistral-7B (used in Historical Event Narrator)
  • LangChain
  • Prompt Engineering
  • LLM-as-Judge (used in Apple)
  • Agentic AI
  • Vector Embeddings (used in Apple)
  • Semantic Search
  • FastAPI (used in PitchPulse, Coding Problem Recommender)
  • REST APIs
  • Docker
  • AWS EC2 / S3 / Lambda
  • Azure ML (used in Coding Problem Recommender)
  • Azure OpenAI
  • Google Vertex AI
  • Firebase (used in BarterBrAIn, PitchPulse, MoodBites)
  • BigQuery
  • JavaScript (used in The Sparks Foundation, Charity Donation Website)
  • Streamlit (used in Historical Event Narrator, Coding Problem Recommender)
  • Flutter (used in BarterBrAIn, PitchPulse)
  • Kotlin (used in MoodBites)
  • Android (used in MoodBites)
  • HTML5
  • CSS3
  • Bootstrap (used in Charity Donation Website)
  • Synthetic Data
  • Retrieval Ranking
  • Offline Eval Pipelines
  • ML Monitoring
  • Model Versioning
  • Observability
  • CI/CD
  • System Design

Bigger bubbles appear in more of my projects and roles. Hover one to see where.

Credentials

Research & certifications

Publications

Conference paper · May 2023

A Comprehensive Review on Optimized Approach for Gastro-Intestinal Tract Segmentation to Improve Cancer Treatment

National Conference on Innovative Computing Research, SVCE, India

Conference paper · Mar 2023

Sentiment Analysis of Twitter Comments – A Survey

National Conference on Youth Advanced Computational Convergence, RGNIYD, India

Certifications

Recommendations

From people I've worked with

Keerthirajan consistently demonstrated a deep understanding of software principles, coupled with a remarkable ability to apply these concepts in practical, real-world situations. I am confident that he will bring the same level of commitment, enthusiasm, and excellence to any role he undertakes. I highly recommend him to anyone seeking a capable and driven professional with a strong foundation in Software Engineering.
Dr. Jayabhaduri Radhakrishnan
Dr. Jayabhaduri RadhakrishnanProfessor
Keerthirajan is one of the top 10 students in our department of Computer Science and Engineering. His strong academic background, along with his capacity to succeed in research and problem-solving, distinguishes him as an extraordinary prospect. I am confident that his ravenous appetite for knowledge, along with his ambition to make a significant contribution, will propel him to unprecedented success in his future endeavors.
Dr. P. Janarthanan
Dr. P. JanarthananMentor & Professor
1 / 2

Open to opportunities

Have a role in mind? Let's talk.

I'm looking for AIML and software engineering roles. Based in San Francisco, CA, open to relocation.

View résumé
keerthirajans@gwu.eduAIML & software engineering opportunities