Skip to content
View hritvikgupta's full-sized avatar
🎯
Focusing
🎯
Focusing

Block or report hritvikgupta

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don't include any personal information such as legal names or email addresses. Markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
hritvikgupta/README.md

πŸ‘‹ Hi, I'm Hritvik Gupta

AI Engineer β€’ LLM Systems β€’ RAG Architect β€’ Full-Stack AI Builder β€’ Founder of MotionNote Β· StoryBorrd Β· Awaken


πŸ› οΈ Tech Stack






🌟 About Me

I’m an AI Engineer specializing in:

  • Large Language Models (LLMs)
  • Retrieval-Augmented Generation (RAG)
  • Voice + Chat AI Agents
  • Vector search (FAISS, LlamaIndex)
  • Multimodal reasoning
  • Distributed Data Engineering (Spark + Delta Lake)
  • Full-stack AI SaaS development

I’ve built:

  • 🎀 AI Voice Chat Agent used by 100,000+ patients
  • 🧠 RAG engines for genomics + scientific research
  • πŸ“ AI Word Processor (MotionNote)
  • 🧩 AI Research Canvas (StoryBorrd)
  • πŸŽ™οΈ AI Avatar Studio (Awaken)
  • πŸ“Š Predictive analytics used by 5000+ employees

🧠 Experience

AI Engineer β€” Penn Medicine (Jul 2024 - Present)

  • Built an AI Voice & Chat system for Perception Care used by 100K+ West Coast patients; designed multilingual speech β†’ RAG β†’ LLM pipeline using LlamaIndex, FAISS, LangChain, FastAPI, and Docker.
  • Engineered data generation for speech and language models retrieval, vector search, and prompt-routing workflows, reducing response latency by ~28% and improving clinical-text retrieval accuracy by 22%.
  • Enhanced PLATLAS genomics platform with ML-based variant ranking and phenotype-similarity scoring; collaborated with Argonne National Lab and leveraged the Aurora supercomputer to run 27M-SNP Nextflow pipelines and added LLM-guided GWAS/EXWAS analysis.
  • Developed PySpark + Delta Lake pipelines standardizing 30+ clinical datasets into OMOP, enabling real-time cohort building and disease-trend dashboards.

Graduate Researcher (NLP) β€” University of California, Riverside (Oct 2022 – Dec 2023)

  • Built large-scale NLP pipelines (Python, Spark, SQL) for 10M+ multilingual research documents, improving tokenization + embedding generation speed by ~40% for downstream scientific analysis.
  • Optimized RAG systems using LlamaIndex + LangChain, raising scientific text retrieval accuracy by 22% and reducing context-window failures through improved chunking + prompt-routing.

Data Analyst β€” Cognizant (Aug 2021 – Aug 2022)

  • Built Python + SQL ETL pipelines processing 20K+ HR, payroll, and marketing records across 50 operational datasets, improving data accuracy and reconciliation reliability by ~35%.
  • Developed Scikit-learn + SAS predictive models for attrition and hiring demand, improving workforce planning accuracy and staffing decisions for 5,000+ employees.

πŸš€ AI Products (Founder)

πŸ“ MotionNote β€” AI Word Processor

The Open Source AI-Powered Word Processor

MotionNote is the fastest and easiest way to create, edit, and enhance documents with AI assistance. It turns text β†’ charts β†’ diagrams β†’ videos using AI.

  • Chart + Diagram Generator: Select text to convert to charts/graphs.
  • Video Generation: Convert notes into high-quality explainer videos.
  • RAG-Powered Contextual Memory: Chat with your documents.
  • AI Copilot: Intelligent writing companion.
  • Tech Stack: FastAPI + Supabase + CopilotKit + Vue 3.

πŸ”— motionnote.com


πŸ“Š MotionExcel β€” AI Data Analyst

Automated Data Analysis & Visualization Dashboard

MotionExcel transforms raw data into actionable insights using AI-driven SQL generation and Python analysis.

  • Text-to-SQL: Query your database using natural language.
  • Automated Dashboards: Instantly generate visual dashboards from your data.
  • Python Data Analysis: Run complex Python analysis scripts on your spreadsheet data.
  • Pivot Tables & Charts: Create pivot tables and charts with a single click.

πŸŽ₯ Demos


🧠 StoryBorrd β€” AI Research Canvas

Figma-style AI canvas for research, storytelling & multimodal workflows.

  • 200+ AI models integrated
  • Graph-memory engine
  • RAG over docs, tables, PDFs
  • Multimodal reasoning
    πŸ”— https://storyborrd.com

πŸŽ™οΈ Awaken β€” AI Avatar Studio

Lifelike AI avatar video generation.

  • Realistic speech-sync avatars
  • Voice cloning
  • Slide / script sync
  • Automated invoice system
    πŸ”— https://awaken.sh

πŸ“š Publications

  • Genome-Wide Pleiotropy Analysis β€” medRxiv, 2025
  • Extractive Text Summarization (ELMo) β€” IEEE I-SMAC
  • EEG Microstate Analysis (RNN) β€” i-PACT
  • LSA Topic Modeling + BERT β€” AI Smart Systems

πŸ† Awards

  • Health-Tech Innovation Accelerator Award – Penn Health-Tech (2025), CIRCA Voice-AI For General Healthcare Services to patients

πŸŽ“ Education

  • Masters In Computer Engineering β€” University of California - Riverside (Sep 2022 – Dec 2023)
  • B.Tech Computer Science β€” GITS Udaipur

πŸ“Š GitHub Analytics

GitHub Stats GitHub Streak
Top Languages


🀝 Connect With Me

πŸ“§ Email: hritvik2920@gmail.com
πŸ”— LinkedIn: https://linkedin.com/in/hritvik-gupta
πŸ’» GitHub: https://github.com/hritvikgupta


πŸ˜„ Fun Fact

Every AI product starts as a tiny script…
and then becomes your entire life πŸ˜‚

Pinned Loading

  1. EEG-DURING-MENTAL-ARTHMETIC-TASK EEG-DURING-MENTAL-ARTHMETIC-TASK Public

    RESEARCH PROJECT, AIM IS TO CLASSIFY THE EEG SIGNALS DURING THE REST STATE AND MENTAL ARITHMETIC TASK STATE

    Jupyter Notebook 1

  2. Image-captioning Image-captioning Public

    Scene understanding, which blends computer vision with natural language processing skills, includes image caption, which automatically generates natural language descriptions based on the content o…

    Jupyter Notebook 1

  3. Docs_classification Docs_classification Public

    This Is NLP base Text Document Classifier Application which first converts the image into text using OCR and then classify it among different sets of classes that text is based upon

    Python

  4. Chatbot Chatbot Public

    ChatBot is the Application of the Natural Language Processing which Trains Neural Networks to produce natural Spoken language.

    Jupyter Notebook 1