Harsh Kumar

Data Science & Machine Learning Enthusiast

Download Resume
Harsh Kumar

Contact Information

Languages

  • Python
  • SQL
  • Java
  • C/C++

Frameworks

  • Pandas, NumPy, Scikit-Learn
  • PyTorch, TensorFlow
  • FastAPI, Flask, LangChain
  • Hugging Face Transformers

Tools

  • Tableau, Linux, MySQL, Postman
  • Docker, AWS, Azure
  • Git, GitHub, Kubernetes
  • Jenkins, Terraform

Soft Skills

  • Problem Solving
  • Critical Thinking
  • Agile/Scrum
  • Teamwork
  • Time Management
  • Technical Documentation

Generative AI & LLMs

  • RAG, Prompt Engineering
  • Vector Database, Deep Learning
  • FAISS, MLOps, Agentic AI

Additional Skills

  • EDA, Data Visualization
  • DevOps, Cloud Deployment
  • Machine Learning, NLP
  • REST APIs

Professional Experience

Product Engineering Intern

December, 2025
  • Developed internal automation scripts to streamline product workflows, reducing manual operational effort by 80% across recurring tasks.
  • Assisted in analyzing product performance data and identifying functional bottlenecks, contributing to more stable and faster releases
  • Collaborated cross-functionally to support day-to-day product operations, ensuring seamless functioning of DishTV’s digital service ecosystem

Software Engineer Intern (DevOps)

Enveu
August, 2025
  • Containerized applications using Docker and orchestrated with Kubernetes, improving deployment consistency across development, testing, and production environments
  • Collaborated with cross-functional teams to optimize infrastructure costs
  • Documented infrastructure architecture and deployment procedures, creating detailed runbooks for team reference

AWS APAC Solutions Architecture Job Simulation

AWS - Forage
April, 2025
  • Designed and simple and scalable hosting architecture based on Elastic Beanstalk for a client experiencing significant growth and slow response times
  • Described my proposed architecture in plain language ensuring my client understood how it works and how costs will be calculated for it

Static Website for Tathagat Tour and Travels

Tathagat Tour and Travels
January, 2025
  • Designed and developed a responsive static website to showcase travel packages, services, and contact information.
  • Implemented clean UI/UX with HTML, CSS, and JavaScript for an engaging user experience.
  • Deployed the website on the web, ensuring fast loading times and accessibility across devices.

Data Science Job Simulation

British Airways - Forage
April, 2024
  • Implemented a simulation underscoring the critical impact of data science on British Airways' operational efficiency, projecting a 10% improvement in customer satisfaction
  • Analyzed 5,000+ customer reviews with Python & SQL, driving targeted improvements and increasing satisfaction by 8%

Data Analytics and Visualization Job Simulation

Accenture North America - Forage
January, 2024
  • Developed a simulation to advise a hypothetical social media client; Leveraged data analytics to boost user engagement by 20% and ad revenue by 12%
  • Processed seven datasets to extract content trends, guiding strategic decisions

Projects

Clause AI: Legal Document Analyzer

April, 2026
  • Built a full-stack legal document analyzer using RAG + GPT-4o that extracts clauses, flags risks, generates plain-English summaries, and compares documents
  • Engineered the backend with FastAPI, Celery, PostgreSQL + PGVector, and FAISS for dual vector storage enabling fast per-document and cross-document search
  • Developed a React + TypeScript frontend with Zustand state management and JWT authentication for secure document upload and interactive Q&A

Cashbit: Personal Finance Tracker

March, 2026
  • Built a full-stack personal finance application with React + Tailwind CSS frontend and Node.js + Express backend, backed by PostgreSQL with Prisma ORM
  • Implemented JWT authentication, category-based transaction management, monthly budgeting, and analytics dashboards with trend visualization
  • Containerized the entire stack with Docker Compose including database migrations, seed data, and Prometheus metrics endpoint

RAG Chatbot

February, 2026
  • Built a production-hardened RAG chatbot with dual reranker support (Cohere API and BGE local model), switchable live in the UI without restart
  • Implemented an 8-stage pipeline: FAISS retrieval, cross-encoder reranking, score normalisation, span-hash synthesis detection, lost-in-the-middle mitigation, and GPT-4o generation
  • Added voice integration (speech-to-text and text-to-speech), analytics dashboard, document upload, and multi-format export (JSON, text, PDF)

WAWC: AWS Well-Architected Watchdog CLI

November, 2025
  • Built a production-ready Python CLI tool that scans AWS accounts for misconfigurations aligned with the AWS Well-Architected Framework
  • Implemented S3 public bucket detection, security group analysis, and RDS backup checks with severity-based findings and remediation guidance
  • Designed for CI/CD integration with non-zero exit codes for findings, JSON export, and support for multi-region scanning and HTML/PDF reports

SignSpeak AI: ASL Gesture Recognition

October, 2025
  • Developed a real-time American Sign Language gesture recognition and translation app using Google MediaPipe hand tracking and React
  • Implemented 7 ASL gesture classifications with confidence thresholds, gesture buffering, and a comprehensive debug panel for troubleshooting
  • Integrated Web Speech API for text-to-speech output with multiple accents, session statistics tracking, and translation history

Certificates

Introduction to Model Context Protocol

Anthropic
August, 2025
  • Acquired in-depth understanding of Anthropic's Model Context Protocol (MCP) for controlling and steering AI model behaviour
  • Learned techniques for providing effective context to large language models to improve response quality and safety
  • Explored practical applications of MCP in developing more reliable and controllable AI systems
  • Gained insights into best practices for prompt engineering and context management in AI applications

Data Visualization Developer Certification

freeCodeCamp
December, 2024
  • Completed 300+ hours of coursework covering responsive data visualization techniques
  • Built interactive and dynamic data-driven visualizations using HTML, SVG, and JavaScript
  • Procured hands-on experience in presenting real-world data insights through visually compelling charts and graphs

Complete Machine Learning & Data Science Program

GeeksForGeeks
June, 2024
  • Foundation for Machine Learning, Deep learning, and Natural Language Processing
  • Acquired a comprehensive understanding of the data life cycle and various stages involved in the data analysis
  • Completed hands-on projects involving EDA and machine-learning techniques

Career Essentials in Generative AI

Microsoft and LinkedIn
September, 2023
  • Gained in-depth knowledge of AI fundamentals, including machine learning, deep learning, and generative models
  • Attained practical knowledge of generative AI applications and their potential across industries
  • Explored real-world use cases of generative AI, including text and image generation models

Programming in Python

Guvi
August, 2023
  • Mastered fundamental Python syntax, proficiently utilising control flow, loops, functions, and data structures
  • Acquired expertise in procedural programming paradigms and associated logical concepts, enhancing capabilities
  • Implemented Python scripts for automation, improving workflow efficiency

Education

Bachelor of Technology

Lovely Professional University | Phagwara, Punjab
August, 2022 - July, 2026

Computer Science and Engineering
CGPA: 8.7

All India Senior School Certificate Examination (CBSE - XII)

Delhi Public School | Gaya, Bihar
April, 2020 - March, 2022

Percentage: 86%

All India Secondary School Examination (CBSE - X)

Takshila School | Gaya, Bihar
April, 2019 - March, 2020

Percentage: 96%