Skip to content

Navigation Menu

Sign in
Sign up
@aarya-pabha
aarya-pabha
Follow

Aarya Mahajan Pabha aarya-pabha

Data Science Graduate , @indiana-university Bloomington

Block or report aarya-pabha

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.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

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

Report abuse
aarya-pabha /README.md

Hi there, I'm Aarya Pabha πŸ‘‹

Research Data Scientist | AI & Machine Learning Engineer

LinkedIn GitHub Email Hugging Face


Typing SVG


πŸ‘¨β€πŸ’» About Me

I am a Research Data Scientist at Indiana University Bloomington and an M.S. in Data Science graduate specializing in production-grade, distributed AI systems, privacy-preserving federated learning, and autonomous multi-agent engines.

  • πŸ”¬ Current Research Focus: Scaling probabilistic ML pipelines across 684-feature matrices and 500+ parallel simulations under extreme low-variance constraints.
  • βš™οΈ Core Expertise: Distributed AI Infrastructure (GCP/AWS), Hybrid RAG with Semantic Routing, Autonomous Agent Networks (BeeAI, LangGraph), Vision Transformers (ViT), and Healthcare MLOps (HIPAA/FHIR).
  • πŸŽ“ Education: M.S. in Data Science (Indiana University Bloomington) | B.Tech in Computer Science & Engineering (VIT).

πŸ› οΈ Tech Stack & Ecosystem

 🧠 Machine Learning & AI :: PyTorch | Scikit-Learn | TensorFlow | Hugging Face | Vision Transformers | OpenCV | XAI
 πŸ€– Agentic & LLM Frameworks :: BeeAI | LangGraph | CrewAI | LangChain | LlamaIndex | Prompt Engineering | Pydantic
 πŸ—„οΈ Databases & Search :: PostgreSQL | MySQL | Redis | Pinecone (Vector DB) | Apache Spark | Firestore
 ☁️ Cloud & Infrastructure :: GCP (Cloud Functions/GCS) | AWS (SageMaker/Timestream/S3/RDS) | Docker | Git | MLflow
 πŸ“Š Analytics & Web Deployment:: Python | R | SQL | PowerBI | Streamlit | Gradio | Hugging Face Spaces | Pandas | NumPy

Tech Stack Icons


πŸš€ Featured Flagship Projects

Project Key Innovation & Architecture Stack & Links
πŸ›οΈ Wall St. AI Analyst Hybrid RAG & Live Market Agent: Built a Semantic Router (OpenAIPydanticProgram) partitioning queries between Pinecone (SEC 10-K RAG with MarkdownElementNodeParser table parsing) and yfinance live API. Evaluated via Ragas framework (83.5% Faithfulness, 91.8% Answer Relevancy). LlamaIndex OpenAI Pinecone Ragas Streamlit
πŸ‘‰ GitHub Repo | 🌐 Hugging Face Space
πŸ›‘οΈ FedMed-XAI Serverless Federated Diagnostics: Engineered GradsSharding architecture on GCP Cloud Functions + GCS (bypassing 50MB+ tensor limits), Zero-Trust Medplum FHIR-native audit logs, and automated PHI scrubbing ETL (PostgreSQL + Redis). GCP Python PostgreSQL Redis Medplum FHIR XAI
πŸ‘‰ GitHub Repo
πŸ€– Autonomous Trading Council Multimodal Multi-Agent Network: Architected Bull, Bear & CIO Agent Council via BeeAI & GPT-4o with cyclic reflection loops, OpenBB v4 & VLMs for chart analysis, and Pydantic schemas (98% pass rate against CIO-Judge checks). BeeAI GPT-4o VLMs OpenBB Pydantic
πŸ‘‰ GitHub Repo
🌲 Wildfire ViT Pipeline Hierarchical Vision Transformer: Built an early wildfire detection pipeline achieving 97.8% binary accuracy across 23 model variations with cascading multi-class classification. ViT PyTorch OpenCV Python
πŸ‘‰ GitHub Repo
πŸ’ Ape Pose Detection Computer Vision Pose Estimation: Engineered deep learning pipeline for accurate primate pose estimation and keypoint joint tracking. PyTorch OpenCV Computer Vision
πŸ‘‰ GitHub Repo
πŸ›οΈ Aspect Sentiment Analysis Web Mining & NLP Engine: Product review extraction pipeline featuring custom spam filtering (20% false positive reduction) and automated analytics integration (40% productivity boost). NLP Scikit-Learn Pandas BeautifulSoup
πŸ‘‰ GitHub Repo

πŸ’Ό Professional Experience Snapshot

  • πŸ”¬ Research Data Scientist | Indiana University Bloomington Sep 2025 - Present
    • Scaled probabilistic machine learning pipelines processing 684-feature matrices across 500+ parallel simulations.
    • Engineered Pandas evaluation scripts filtering 22,000+ false positives down to 467 actionable targets.
  • 🏒 AI Architect Intern | Monage Corporation Jun 2025 - Aug 2025
    • Architected predictive healthcare platform targeting 1M active users (<500ms latency) via AWS SageMaker, Timestream, RDS PostgreSQL & S3.
    • Designed HIPAA-compliant AI integration workflows for US/India MVP inference routing.
  • πŸ’» Data Scientist Intern | Wipro Limited May 2023 - Jul 2023
    • Built AI troubleshooting prototype integrating GPT-3 with Tesseract OCR; fine-tuned Stable Diffusion v1.5 via DreamBooth on curated datasets.

πŸ“œ Certifications & Education

  • πŸŽ“ M.S. in Data Science β€” Indiana University Bloomington
  • πŸŽ“ B.Tech in Computer Science & Engineering β€” Vellore Institute of Technology
  • πŸ† IBM Certified: RAG and Agentic AI
  • πŸ† DataCamp Certified: SQL Associate

🀝 Let's Connect & Collaborate!

Feel free to reach out for research collaborations, AI/ML engineering projects, or technical inquiries.

LinkedIn Email Hugging Face

Pinned Loading

  1. Wildfire-Detection-using-Vision-Transformers Wildfire-Detection-using-Vision-Transformers Public

    Hierarchical Vision Transformer pipeline for early wildfire detection β€” 97.8% binary accuracy, ablation study across 23 model variations

    Jupyter Notebook 2

  2. Ape-Pose-Detection Ape-Pose-Detection Public

    Jupyter Notebook

  3. Aspect-Based-Sentiment-Analysis Aspect-Based-Sentiment-Analysis Public

    Jupyter Notebook

  4. financial-ai-agent financial-ai-agent Public

    Python

AltStyle γ«γ‚ˆγ£γ¦ε€‰ζ›γ•γ‚ŒγŸγƒšγƒΌγ‚Έ (->γ‚ͺγƒͺγ‚ΈγƒŠγƒ«) /