AI & Data Science Professional

I build production ML systems — forecasting, knowledge graphs, and edge vision.

From messy data to deployed models: I ship analytics, GenAI retrieval, and real-time computer vision.

>

Sarthak Kapaliya

Sarthak Kapaliya

AI, Data Science, MLOps, Forecasting, Knowledge Graphs.

Experience0 roles
Projects0
Publications0

About

Who I am

Hi 👋! I'm Sarthak Kapaliya.

I am an AI and Data Science professional focused on ML, NLP, computer vision, forecasting, and production-grade MLOps. I enjoy building systems that are explainable, reliable, and measurable. My work bridges the gap between research and real-world deployment.

Recently, I’ve shipped production workflows in insurance analytics (variance analysis, forecasting, knowledge graphs), edge AI for real-time vision, and GenAI-driven document intelligence for complex financial data.

Highlights

Impact
  • Mortality experience analytics & forecasting (ARIMA/Prophet) with MLflow and Power BI storytelling.
  • Real-time people counting on 200+ edge devices; Jetson Nano + DeepStream; DVC + MLflow for CV models.
  • Knowledge graph (Neo4j/Cypher) for GenAI retrieval, boosting chatbot accuracy by 35%.
  • LLM-enabled document intelligence; confidence routing + human-in-the-loop for noisy tax PDFs.
  • Multimodal AI mock-interview platform with speech-to-text + FER; validated with 40+ users.

Experience

Work Experience

May 2025 – Dec 2025

Data Science Intern (Group Function AI)

Manulife Insurance · Toronto, Canada

  • Automated mortality experience variance/trend analysis with SQL & PySpark; reduced manual workflows by 40% via Power BI + AI narratives.
  • Built ARIMA & Prophet cashflow forecasts, improving MAPE by 15% for proactive financial decisions.
  • Production ML with MLflow for experiment tracking, reproducibility, and CI/CD-aligned deployments.
  • Engineered Neo4j treaty knowledge graph and multi-hop reasoning to support GenAI retrieval; improved chatbot accuracy by 35%.
Jan 2024 – Aug 2024

AI Trainee

Buze Platforms Private Limited · Ahmedabad, India

  • Election monitoring with real-time people counting; managed 200 AWS Linux servers with edge AI cameras; +60% accuracy.
  • Real-time human detection on Jetson Nano + DeepStream; +50% processing efficiency.
  • Helmet violation detection (transformer-based) on AWS EC2 with DVC + MLflow; +55% speed, +30% accuracy.
Jun 2023 – Aug 2023

Data Science Intern

Techdome Solutions Private Limited · Bhopal, India

  • Chat toxicity analyzer boosting interaction by 25% (hate/obscenity detection).
  • Topic-model recommendation engine improving UX by 30% through user-interest clustering.
May 2023 – Jul 2023

Research Intern

Indian Meteorological Department · Bhopal, India

  • Advanced ML on Doppler radar data; improved lightning detection accuracy by 20%.
  • Enabled early weather warnings through better thunderstorm visualization.
Jun 2023 – Jul 2023

Project Trainee

Neurapses Technology · Pune, India

  • Generative AI for MongoDB query synthesis to boost developer productivity.
  • Visualization layer for automated query insights.
May 2022 – Jul 2022

Project Trainee

Semiconductor Laboratory (MeitY) · Chandigarh, India

  • Computer vision surveillance to classify 7 vehicle types; persisted counts for analytics.

Projects

Featured Work

Intelligent Tax Document Extraction System

2025
Azure Document Intelligence LLM prompts HITL

Noisy, multi-format tax PDF processing with confidence routing and human-in-the-loop; cut manual review time by 80%.

AI-Powered Mock Interview Platform

2025
LLMs STT FER UX Research

Multimodal system giving real-time feedback; 40+ user study showed 25% improvement in response quality.

Plastic Waste Detection

Jun – Sep 2023
Python Flask Roboflow Ultralytics

End-to-end detection pipeline for plastic waste with hosted inference.

Song Recommendation Engine

Apr – May 2023
Python Flask NLTK ML

NLP + topic modeling to personalize music discovery.

Archive

More Work

OCR Web Portal

Flask · Azure · JS · Oct – Dec 2022

PRINTF — Printing Service Web Portal

Node.js · Firebase · React · Twilio · Feb – Mar 2023

Academics

Education

Master of Engineering, Computing and Software

Sep 2024 – Apr 2026

McMaster University · Hamilton, Ontario, Canada

Bachelor of Technology, Computer Engineering

Nov 2020 – Jun 2024

Pandit Deendayal Energy University · India

Leadership

Positions of Responsibility

Student Placement Coordinator

Career Development Cell, PDEU | Jan 2023 - Present

AI/ML Core Member

Encode: CSE Club of PDEU | Nov 2022 - May 2023

Public Relation Head

IEEE Student Chapter, PDEU | Jul 2021 - May 2023

Social Media Head

Civiqueniti, Public Administration Club of PDEU | Jul 2022 - May 2023

Research

Publications

(Journal) Facial Emotion Recognition With Deep Neural Network - Accepted

(Journal) AI Enabled Ozone Forecasting Model using DNN for Indian Air Quality Data - In Review

(Journal) Water Quality Prediction Using Machine Learning - In Review

(Journal) An Intelligent Classification of Human Protein-Coding Genes: An AI-Driven Approach - In Review

(Journal) Advancement in Efficient Approaches for Detection of Prevalent Phishing Attacks - In Review

Toolkit

Skills

PythonPython
JavaJava
RR
MySQLMySQL
PyTorchPyTorch
TensorFlowTensorFlow
scikit-learnscikit-learn
PandasPandas
NumPyNumPy
PySparkPySpark
MLflowMLflow
AnacondaAnaconda
FastAPIFastAPI
MongoDBMongoDB
PostgreSQLPostgreSQL
DockerDocker
KubernetesKubernetes
AWSAWS
AzureAzure
GCPGCP
VS CodeVS Code
GitGit
LinuxLinux

Certifications

Creds

Udacity: Nanodegree in AI Programming with Python (Valid until: Dec 2026)

Microsoft: Certified Azure AI Fundamentals (Valid until: Nov 2026)

NVIDIA: Fundamentals of Deep Learning; AI for Predictive Maintenance; Transformer-based NLP Apps (Valid until: Oct 2026)

NPTEL IIT Kharagpur: Deep Learning

IBM: Data Analytics Internship at CSRBox; Introduction to Data Science; Data Science Tools

ISRO: Geoprocessing With Python Certification

Columbia University (edX): Machine Learning for Data Science and Analytics

Udemy: Complete ML & Data Science with Python; Python for Beginners (2022)

Lab notebook

Latest Writing

View all posts

Loading latest posts…

Contact

Let’s build something