Skip to content
View RaySatish's full-sized avatar

Highlights

  • Pro

Block or report RaySatish

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 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
RaySatish/README.md

Satish Prem Anand

Typing SVG



LinkedIn Gmail Codolio GitHub


Profile Views


$ whoami

→ Pre-final year @ VIT Vellore · M.Sc. Computational Statistics & Data Analytics · 9.23 GPA
→ Building production ML systems — not notebooks, not demos
→ Quantitative finance: portfolio optimisation, market microstructure, risk modelling
→ Agentic AI: LangGraph systems, autonomous workflows
→ Data engineering: distributed pipelines with Spark, Kafka, Hadoop on AWS
→ 546 DSA problems solved across platforms
→ 2 conference papers · 1 research internship (IIIT Allahabad)
→ Available immediately (remote) · Mid-May (on-site)

📊 GitHub Stats

  








🧩 DSA Stats



Platform Total 🟢 Easy 🟡 Medium 🔴 Hard Link
LeetCode
TakeUForward
GeeksForGeeks
✨ Combined

🏆 TakeUForward Sheet Progress

Sheet Progress %
📘 A2Z Sheet 74%
🔁 DSA Concept Revision 73%
⚡ DSA Quick Revision 63%
👁 Blind 75 48%
🎯 Striver 79 68%
🗂 SDE Sheet 47%

🔗 TakeUForward Topic Mastery

Topic % Mastery
🔗 Linked Lists 100%
🌲 Trees 91%
🕸 Graphs 89%
🔍 Binary Search 85%
🧮 Dynamic Programming 27%

🚀 Featured Projects

Fully agentic job application pipeline. Parses JDs with NLP-based NER, retrieves CV sections via RAG, scores fit with sentence transformers, tailors LaTeX resumes, drafts emails — fires applications with a single Telegram tap. Self-improving match scorer retrains on personal outcomes.

LangGraph LiteLLM RAG ChromaDB Supabase Playwright Telegram Gmail MCP

Distributed pipeline to detect wash trading, spoofing, and pump-and-dump in live Binance data. Fault-tolerant ETL on AWS EMR with idempotent jobs, dead-letter queues, and real-time dashboard on EC2.

Apache Spark Kafka Hadoop AWS EMR Binance API ETL

Per-ticker LSTM forecasting with FinBERT sentiment analysis and Monte Carlo Sharpe ratio optimisation. Backtested across 10 stocks — sentiment-augmented model validated with paired t-test and Wilcoxon testing. Presented at 26th SET Conference.

LSTM FinBERT yfinance Monte Carlo Streamlit Significance Testing

Self-pruning network with learnable sigmoid gates and L1 sparsity regularization on CIFAR-10. Achieves 277× compression — 99.6% of weights pruned while improving accuracy by 2% over baseline. Custom PrunableLinear layer, per-layer sparsity analysis, and 17 auto-generated visualizations.

PyTorch Neural Pruning CIFAR-10 L1 Regularization MPS Accelerated matplotlib


🛠 Stack

Languages

Python SQL R Java MATLAB

ML / AI

PyTorch TensorFlow scikit-learn LangChain ChromaDB HuggingFace LiteLLM

Data Engineering

Apache Spark Kafka Hadoop PostgreSQL Supabase Pandas

Cloud & DevOps

AWS Docker FastAPI Git Linux Streamlit


Satish Prem Anand · VIT Vellore · Pre-final year
Available immediately (remote) · Mid-May (on-site)

Email

Pinned Loading

  1. CouchHire CouchHire Public

    Job applications, automated. You stay on the couch.

    Python 2

  2. Market-Surveillance-System Market-Surveillance-System Public

    Big-data pipeline detecting wash trading and pump & dump in trade data using Apache Spark & Kafka, deployed via Streamlit.

    Python 5

  3. self-pruning-neural-network self-pruning-neural-network Public

    A feed-forward neural network that learns to prune itself during training using differentiable gate parameters

    Python 1

  4. aritra0309/nifty50-portfolio-optimizer aritra0309/nifty50-portfolio-optimizer Public

    Nifty 50 portfolio optimizer

    Python 1