Sirshendu
is a Data Scientist at Indicator Lab, specializing in developing a data-centric AI aggregator tool for investment strategy and risk mitigation. With expertise in data science and AI, he applies techniques such as Time-Series Analysis, Graph Analysis, and Deep Learning to address challenges in finance.
He holds a MS in Data Science from Worcester Polytechnic Institute (MA, USA), focusing his thesis on an anti-money laundering tool using graph neural networks.
Previously, at Tata Consultancy Services, he managed servers for a US bank.
Sirshendu is dedicated to continuous learning in data science and AI.
EXPERIENCE
Data Scientist
Indicator Lab
- • Developed data-centric AI aggregator tools for investment strategy and risk mitigation
- • Applied advanced techniques including Time-Series Analysis, Graph Analysis, and Deep Learning
- • Built predictive models for financial market analysis and portfolio optimization
- • Collaborated with cross-functional teams to deliver AI-powered financial solutions
Research Assistant
Worcester Polytechnic Institute
- • Developed a novel fraud detection approach using Graph Neural Networks (GNN) at Dr. Fabricio Murai's lab. The work resulted to my MS thesis.
- • Implemented a ML-based ransomware defense system using file-system activity monitoring and low-level I/O packet sniffing in Dr. Lorenzo De Carli's lab.
- • Proposed a moving-target defense for IoT security while working in Dr. Lorenzo De Carli's lab, injecting syntax mutation to disrupt software homogeneity, with work published at EAI SecureComm 2022.
Assistant System Engineer
Tata Consultancy Services
- • Maintained a segment of ETL pipeline for a US-based Bank client, ensuring efficient and secure data processing.
- • Implemented scripts to generate graphs depicting changes in data flow.
Computer Science online instructor
mycareerwise
- • Advised by Dr. Debashis Roy, founder of MyCareerwise.
- • Curated content useful for cracking technical job interviews, and entrance exams such as GATE, NET, etc.
PROJECTS
A rewardable and sensible oracle that automatically finds market demand for data and motivates developers to upload needed data streams with AI integration.
🏆 2nd Place Winner - OPL x Sei Hackathon 2023
AI agent system based on Model Context Protocol (MCP) using LangGraph's ReAct agent to analyze Polymarket's real-world event-based trading options with integrated API data and trend analysis.
🤖 Model Context Protocol & Agentic AI
Retrieval-Augmented Generation pipeline to fetch, embed, and analyze cryptocurrency news using vector embeddings and contextual querying with properly cited sources.
📰 Retrieval-Augmented Generation (RAG)
Fine-tuned FLAN-T5 model for Indian news summarization using LoRA technique, achieving 34% improvement in ROUGE-2 scores with enhanced fluency and context awareness.
📊 34% ROUGE-2 improvement • Parameter-Efficient Fine-tuning
Master's thesis project developing an advanced AML detection system using graph neural networks to identify suspicious financial transactions and patterns.
📚 MS Thesis Project - Worcester Polytechnic Institute
GALLERY
CONTACT
Get In Touch
I'm always interested in discussing new opportunities, collaborations, or innovative projects in data science and AI. Feel free to reach out!