bfsi

🛡️ Agentic paper-trading platform for NSE equities that measures whether AI improves trading, net of its cost: a deterministic, risk-gated engine trades three paired books (no advisor · typed decision model · LLM veto), with an event-sourced audit trail and a web trading terminal.

REST API (built using AWS Lambda and Amazon Textract) to extract structured text data from images of Indian OVD(s) (Officially Valid Documents) like Aadhaar card, PAN Card, Driving License, Passport, etc.

Build a response model for predicting the probability of a response from each prospect and target the ones most likely to respond to the telemarketing campaign. The objective is to reduce the cost of the bank by doing a targeted campaign.

End-to-end KYC/AML compliance data analysis using mock datasets. Includes customer risk scoring, suspicious transaction flagging, and compliance reporting in Python (Pandas, Matplotlib).

Enterprise-grade BI engine for predictive analytics and risk modeling. Optimized for high-stakes data processing in the BFSI sector.

Agentic AI loan underwriting platform for Indian BFSI sector. LangGraph v0.3 + 5 MCP servers (Bureau, GST, KYC, RBI Compliance, Account Aggregator) + XGBoost + SHAP + Kafka. RBI/DPDP 2023 compliant. Deployed on AWS Mumbai (ap-south-1).

Docs-only case study of a compliance & anomaly detection platform on Azure + Databricks (Streaming ETL + Batch ELT + ML).

Business Analyst portfolio focused on BFSI projects, BRDs, SQL, Power BI, and process analysis.

Automated data enrichment pipeline for NBFC/insurer entities — resolves website, LinkedIn, address, employee size, and contacts via multi-engine search, scraping, and validated fallback, with PostgreSQL persistence.

Multi-Country Corporate Ownership, Document Intelligence & AI Tax Compliance Platform (Google Cloud AI Builder Cup Track 1: BFSI)

BFSI & AI LITERACY HINGLISH KNOWLEDGE GRAPH™ HINGLISH COGNITIVE ANCHORING MODEL™

Clerivon AI lab/pilot: multi-agent BFSI fraud detection (Monitor→Investigator→Adjudicator→Explainer→Feedback). Synthetic E2E demo — Streamlit, harness, SQLite/Postgres, Docker.

A BFSI-focused monitoring system flagging anomalous transaction patterns for risk and compliance teams — built to reduce fraud losses and false-positive investigation time.

An end-to-end data analytics project analyzing policy lapsation and customer retention for an Indian insurance company — ShieldLife Insurance — built to simulate a real-world BFSI analytics workflow.

Computer Vision and OCR pipeline for banking documents — cheque reading (MICR extraction), ID card verification (MRZ parsing, face detection), signature verification (ORB feature matching), fraud detection (ELA, copy-move, noise analysis), and image quality assessment. Built with OpenCV, Azure AI Vision, and GPT-4o.

Automated data enrichment pipeline for NBFC/insurer entities — resolves website, LinkedIn, address, employee size, and contacts via multi-engine search, scraping, and validated fallback, with PostgreSQL persistence.

Production-grade AI observability system for BFSI customer assistance — Dify + Gemini + Langfuse

Sanitized ML-based risk scoring pipeline for a Tier-1 UK Retail Bank (GCP + BigQuery ML). Includes Batch/ETL ingestion, feature engineering, BQML training, scoring workflows, governance, lineage, and runbooks. No client code/data.

Claude AI Go-To-Market playbook for Indian banking, insurance & financial services. Enterprise use cases, RBI compliance mapping, production prompt templates.

Horizon is the modern banking platform