Hi, I'm
Shankar
Deenadayalan
AI Engineer & Data Scientist — I ship models where the business actually lives.
Getting started
- Double-click My Projects for live GitHub repos
- Open Resume.pdf in the Reader — right-click it for Properties
- Contact to hire me before the next reboot
Shankar Deenadayalan San Diego, CA · MS Business Analytics @ UCSD (Rady Merit Scholar) Four years turning messy business problems into shipped ML: * Cart-abandonment scoring engine (XGBoost) that recovers $58K in revenue every quarter. * Geospatial ML at BCG that helped a $50M ride-sharing fleet reach 50% electrification. * An investment recommender that lifted engagement 31% and portfolio diversification 24%. * ETL and dimensional models saving 20 hrs/week across 15+ stakeholders. Currently training at UCSD — customer analytics, recommender systems, predictive modeling. I like models that meet the business where it lives, and personal sites that boot like it's 2003.
Shankar XP
AI Engineer Edition
Version 2026
AI Engineer Edition
Version 2026
Registered to:Your open ML role
Processor:4 yrs field experience @ 95% Python
Memory:$58K/qtr · $50M fleet · +31% engagement
Stack:Python · SQL · PyTorch · sklearn · LangChain · RAG · Tableau
Gramercy Funds Management — AI CapstoneApr–Jun 2026
- Agent-operable analytics platform (Python) powering a multi-factor country-risk index — AI agents refresh, re-weight and extend it with no human intervention beyond configuration
- Modular connectors ingesting REST APIs, CSV/Excel and unstructured PDFs into one queryable schema with full data provenance
- NLP workstream scoring thousands of earnings-call transcripts (TF-IDF, transparent rule-based scoring)
Kevali — Analytics ManagerOct 2024 – Jun 2025
- Cart-abandonment prediction model (XGBoost) recovering 21% of abandoned carts — $58K in quarterly revenue
- ETL pipelines + dimensional models automating reporting for 15+ stakeholders: −60% ad-hoc queries, 20 hrs/week saved
- +15% conversion via regression modeling, A/B testing and cohort analysis
Tap Invest — Senior Business AnalystMar–Sep 2024
- Investment recommendation engine (collaborative filtering): +24% portfolio diversification, +31% engagement
- Investment-behavior models surfacing proactive trend insights — monthly reporting close time −25%
- Analyzed 200+ leasing contracts: +15% underwriting accuracy, assessment time 3 days → 1 day
Boston Consulting Group — Research AssociateJun 2021 – Sep 2023
- Route planning + demand forecasting (Python, geospatial) enabling a $50M ride-sharing fleet to electrify 50%
- Mapped 6,500+ charging outlets into a predictive market model for India's EV ecosystem
- Segmented 10,000+ users into 3 high-value segments — +8% projected revenue
UC San Diego — MS Business AnalyticsJul 2025 – Dec 2026
- Rady School of Management Merit Scholarship
- Customer analytics, recommender systems, predictive modeling
- ★ NOW RUNNING
Ramaiah Institute of Technology — BE Mechanical Engineering2017–21
To:sdeenadayalan@ucsd.edu
Subject:Let's ship something real
Response time: faster than XP boots on period hardware.
Crossings: 0
Click characters to board the boat. Monsters must never outnumber humans.
Microsoft Windows XP [Version 5.1.2600] (C) Copyright 1985-2003 Deenadayalan Corp. C:\Documents and Settings\Shankar>shankar --stats
ShankarXP is ready
Double-click the desktop icons to explore projects, resume and more.