About

Hello, I’m Kaxit Pandya, a passionate researcher, engineer, and problem solver with a strong focus on Artificial Intelligence (AI) and Machine Learning (ML). I’m currently pursuing my Master in Computer Engineering specializing in AI at the University of Waterloo, where I’m exploring Generative AI (GenAI) and its integration into real-world applications

With experience as a Research Assistant at leading institutions like Indian Space Research Organisation (ISRO), Indian Institute of Technology (IIT) Gandhinagar, National Chung Cheng University in Taiwan, and University of Waterloo, I’ve tackled diverse challenges ranging from deep learning models for radar pulse classification to federated learning for water quality prediction.

Professionally, I work as an AI Engineer at Defai Digital, where I engineered a GraphRAG pipeline integrating Memgraph with Langchain to enable structured knowledge graph retrieval and contextual reasoning. I’ve also built LLM-powered chatbots, fine-tuned YOLO for document segmentation, and orchestrated complex workflows with multi-agent systems. Previously, as an AI/ML Data Engineer at Genellipse Inc., I developed OCR pipelines using Google Cloud Vision API, applied BERT-based models to extract data from U.S. death certificates, and implemented LLMs for deriving actionable insights from complex data structures.

  • Birthday: 23 March 2002
  • Phone: +1 365 998 6883
  • City: Toronto, Canada
  • Email: kaxitpandya@gmail.com

Interests

Gen AI

Machine Learning

Artificial Intelligence

Data Engineer

Education

Masters in Computer Engineering (AI Specialization)

August 2024 - August 2025
CGPA: 8.6
Relevant Coursework:
  • Generative AI
  • Large Language Models (LLMs)
  • Machine Learning

B.Tech in Computer Engineering

June 2020 - July 2024
CGPA: 9.6
Relevant Coursework:
  • Artificial Intelligence
  • Machine Learning
  • Data Structures & Algorithms

Research Experience

An Intelligent Fashion Object Classification Using CNN (Published)

Federated Ensemble based Water Quality Classification on Drone Images (Submitted)

AI-Based Water Quality Prediction Using Machine Learning and Synthetic Minority Oversampling Technique (Accepted)

Advancement in phishing attacks and Machine learning approaches (Submitted)

Online Certification

‐ ‘AWS Certified Machine Learning Engineer Associate’ by AWS

‐ ‘LangChain Mastery’ and ‘OpenAI Python API Bootcamp’ through Udemy

‐ ‘GenAI with LLMs’, ‘Machine Learning Specialization’, and ‘Deep Learning Specialization’ from DeepLearning.AI

‐ ‘Data Analyst with Python’ from IIT Madras through NPTEL

‐ ‘The Web Developer Bootcamp 2023’ through Udemy

Experience

Defai Digital

AI Engineer

January 2025 - Present
  • Engineered a GraphRAG pipeline by integrating Memgraph with Langchain, transforming unstructured data into structured knowledge graphs for improved multi-hop reasoning and contextual retrieval
  • Developed AI chatbots using LLM-based agents to automate user interactions, fine-tuned YOLO for precise document segmentation, and integrated agent frameworks to orchestrate complex multi-step workflows

Genellipse Inc.

AI/ML Data Engineer

October 2024 - December 2024
  • Built a text extraction pipeline for processing diverse document formats (PDF, Word, Excel), leveraging Google Cloud Vision API to extract data from U.S. death certificates with high accuracy
  • Utilized BERT-based AI models to map key-value pairs from U.S. death certificate fields accurately and implemented large language models (LLMs) to interpret extracted information, delivering actionable insights for enhanced client reporting and decision-making

Dr. David Hammond, University of Waterloo

Research Assistantships (RA)

September 2024
  • Implemented an automated web scraping system using Selenium to extract detailed information on grocery items from Walmart, Sobeys, and Target websites, with integrated data export to Excel sheets for efficient analysis
  • Leveraged an image download module to systematically retrieve product images from scraped websites and utilized advanced OCR (Optical Character Recognition) techniques to extract textual information from these downloaded product images

Indian Institute of Technology (IIT) – Gandhinagar

Research Intern

June 2024 - August 2024
  • Designed a Deep Learning model using Generative Adversarial Networks (GANs) to perform inverse design of 2D heterostructure materials, enabling efficient exploration of material properties
  • Conducted post-processing of atomistic simulation data using Python to extract and visualize material properties
  • Converted QuantumATK software functionalities into Python scripts to automate the creation of various heterostructure configurations

TEEP@ASIA PLUS PROGRAM at National Chung Cheng University, Taiwan

Research Intern

January 2024 - May 2024
  • Implemented Federated Learning (FL) model to predict water quality parameters using drone-captured images, ensuring privacy across geographically distributed datasets, and authored a Research Paper detailing the methodologies and results.
  • Executed robust aggregation strategies such as FedAvg, FedAvgM, FedMedian, and FedProx using the Flower Framework, while mastering key Federated Learning (FL) concepts, including handling non-IID data distributions and managing system heterogeneity.
  • Applied Explainable AI (XAI) techniques, such as SHAP, to interpret predictions, highlighting parameter significance and enhancing transparency and trust.

Space Application Center, Indian Space Research Organisation (ISRO)

Research Intern

May 2023 - July 2023
  • Developed Deep Neural Networks for predicting classification of Pulse Radar Interval Modulation
  • Processed and handled noisy, raw radar pulse data directly from ISRO’s satellites
  • Achieved 94% classification accuracy in identifying complex radar Pulse Repetition Intervals (PRI) patterns, including staggered, jittered, dwell and switch, sliding, linear, and sinusoidal

HumanAnsys

Software Developer Intern

Jan 2023 - March 2023
  • Understanding Cloud systems, architecture, and AWS services
  • Using AWS services (EC2, S3 bucket, Lambda Functions) to deploy a video calling website

Hybrid Utopia

Web Developer Intern

June 2022 - Dec 2022
  • Create websites from scratch using several modern-day frameworks like node, express
  • Performing Create, Read, Update, Delete (CRUD) operations using MongoDB

Projects

Q&A Application on Private Documents

Designed a Retrieval-Augmented Generation system for querying custom PDF and document formats.

AI-Powered Resume Analyzer

AI-powered Resume Analysis System

AI-Powered Instagram Caption Generator

Developed using LangChain and Google’s PaLM API to create contextually relevant captions from user inputs.

Custom ChatGPT

Built a custom conversational app with LangChain and Streamlit, replicating a ChatGPT-like user experience.

Fine-Tuning LLMs on Fleet Maintenance Data

Fine-tuned Large Language Models (LLMs) for predictive Fleet maintenance using PEFT techniques like LoRA

Summarizing Large Documents

Implemented summarization techniques with LangChain and OpenAI models to process and summarize large documents.

DeepFake Audio Detection System

Preprocessed audio data with Librosa and classified audio as either Deepfake or human-generated using pre-trained models.

Above Ground Biomass Prediction

Above Ground Biomass Prediction

Performed hyperparameter optimization on a sparse autoencoder, extracted latent-space embeddings, and used them as inputs to hyperparameter-tuned neural network and random forest

Water Quality Index Prediction

Developed and deployed predictive models using various machine learning algorithms for water quality prediction.

Inventory Management System

Built a CRUD-based Inventory Management System using Node, Express, MongoDB, and Tailwind CSS.

Chrome Extension for Tracking Links

Developed a Chrome extension that saves current tab URLs and user notes for easy access and management.

Skills

Languages

Python C C++ JavaScript SQL

Libraries

NumPy Pandas Matplotlib TensorFlow LangChain AWS SageMaker Selenium

Framework

Tailwind CSS Node.js Express

Database

MongoDB MySQL

Add-on Skills

Tableau

Extra-Curricular Activities

Technical Project Manager | Waterloo AI Institute (WAT.AI) | University of Waterloo

Sep’24 – Sep’25

  • Building deployable AI models on onboard satellites, focusing on the semantic segmentation of methane plumes using hyperspectral ML models
  • Leading AI research projects, managing teams, ensuring timely completion and quality deliverables in collaboration with industry partners

Engineering Representative | Graduate Student Endowment Fund (GSEF) | University of Waterloo

Sep’24 – Sep’25

  • Reviewing and evaluating funding applications, ensuring strategic allocation of university resources across diverse academic disciplines to graduate student projects and initiatives
  • Providing strategic input on funding decisions to enhance the academic and professional development of graduate students

Chairperson | GeeksForGeeks - GeeksForGeeks Student Chapter of PDEU | Gandhinagar, Gujarat

June’22 – July’23

  • Management of Events and Workshop at regular intervals
  • Leading a team of 50 members and Developing Coding Culture at our University

AWS DeepRacer Contributor | Amazon Web Service DeepRacer

August‘22

  • Participated in AWS Deepracer Student league and came in the top 10% of racers out of 4000 students
  • Made a Reward function for training DeepRacer Car on Reinforcement Learning Model using AWS

Web Development Head | SnT - Science & Technical Club of PDEU | Gandhinagar, Gujarat

June’22 – July’23

  • Created a Website for the university flagship event named Tesseract (GitHub)

Open Source Contributor | Hacktoberfest 2022

October‘22

  • Made more than 6 contributions to open source via GitHub using HTML, CSS, JavaScript, Tailwind CSS, Node.js, MongoDB

Contact

My Address

Toronto, Ontario, Canada

Social Profiles

Email

kaxitpandya@gmail.com

Contact

+1 365 998 6883