Sanket Chavan — Senior Data Analyst

WHO I AM

About Me

I'm Sanket — a data engineer and analyst who genuinely enjoys turning messy, complex data into something a business can actually act on. I hold a Master of Science in Data Science from the University at Buffalo, and I've spent the last 5 years at Deutsche Bank building production-grade compliance data systems that real teams depend on every day.

At Deutsche Bank, I serve as Client Service Manager (CSM-1) for two enterprise compliance applications, owning them end-to-end across APAC, Europe, and the Americas. My work spans everything from building real-time ETL pipelines that monitor 131 daily broker feeds, to designing Power BI dashboards used by compliance officers and executive stakeholders, to engineering the automation frameworks that eliminated an entire category of manual work. I'm also the primary bridge between Compliance Officers and the Engineering team — translating regulatory requirements into data architecture and delivering ad-hoc analysis when it matters most.

Outside of work, I don't switch off. I independently designed and built MktMoves — a production-grade SEC filing intelligence platform that processes 2.6M+ institutional investment filings through a fault-tolerant Airflow pipeline and surfaces them through a clean Next.js frontend. It's the kind of financial intelligence that was previously only accessible through expensive data vendors. I built it solo, end-to-end, because I wanted to and because I could.

I'm currently relocating to Austin, TX, and looking for my next challenge as a Senior Data Analyst, Business Intelligence Engineer, or Data Engineer. Long-term, my goal is to run a team of data enthusiasts who build the kind of data infrastructure and products that drive real business outcomes — not just reports nobody reads.

When I'm not deep in a SQL query or debugging a pipeline at midnight, you'll find me on a soccer field, gaming, planning the next road trip, or unwinding somewhere near the water. I believe the same qualities that make a good data person — curiosity, attention to detail, and comfort with ambiguity — also make for pretty good company.

WHAT I USE

Skills

Languages & Querying SQL, Python, Shell Scripting, R, PySpark

Databases & Cloud Oracle, PostgreSQL, MongoDB, MS-SQL, GCP, AWS

Data Engineering & ETL Apache Airflow, Control-M, Kafka, FastAPI, Spark

Data Modeling Star Schema, Materialized Views, Indexing, DAGs, Normalization, Workflow Automation

BI & Visualization Power BI, Tableau, Matplotlib, Plotly, Pandas, NumPy

ML & Analytics Regression, Classification, NLP, RAG, Scikit-learn, NLTK

Other Tools Ansible, Docker, Git, ServiceNow, JIRA, Postman, Jupyter

WHERE I'VE WORKED

Experience

Deutsche Bank

Jan 2021 – Present

Senior Data Analyst

Employee Trading Compliance Applications

SQL Python Power BI GCP

As Client Service Manager (CSM-1), owned full end-to-end operational responsibility for two enterprise compliance applications, overseeing broker onboarding, infrastructure changes, capacity planning, and release management across APAC, Europe, and Americas.
Engineered a real-time broker feed monitoring system using Control-M job scheduling, Oracle SQL validation, and hash-based integrity checks to ingest and validate 131 daily feeds from 52 brokers. Integrated automated incident creation via ServiceNow APIs, cutting resolution time from 4–5 days to under 2 hours and eliminating 100% of manual tracking previously flagged in regulatory audits.
Designed and deployed an automated reporting pipeline using Shell scripting and Oracle stored procedures, eliminating manual generation of 150+ weekly compliance reports, reducing L2 workload by 73% and report delivery time by 85%, scaled bank-wide across 18+ applications.
Replaced manual SSH-based production access with an Ansible playbook framework implementing RBAC, multi-level approvals, and full execution logging, achieving 100% elimination of privileged logins and full audit traceability across all compliance applications.
Built a Power BI analytics platform on top of ServiceNow Oracle data using Materialized Views, incremental refresh, and optimized SQL indexing, delivering 13 KPIs that produced a 400% improvement in L2 resolution time (5 days → 24 hours), adopted org-wide.
Acted as the technical bridge between Compliance Officers and the Engineering team, translating regulatory requirements into data architecture decisions and delivering ad-hoc analysis to support executive and audit stakeholders.
Led enablement of a 6-person L2 support team across 3 regions: authored KB articles, conducted KT sessions, and tracked performance via live dashboards, sustaining resolution time improvements and reducing repeated incidents.
Currently contributing to enterprise GCP migration, working hands-on with BigQuery, GKE, and Cloud Storage to modernize compliance data infrastructure from on-prem Oracle to cloud-native architecture.

People Tech Group

Sep 2019 – Nov 2020

Junior Data Scientist

Elliptica Data Platform

Python Kafka SQL Tableau MongoDB MS-SQL

Engineered fault-tolerant data connectors for relational and non-relational databases to capture real-time changes using Python and Kafka.
Tracked and streamed database changes to Tableau dashboards through Redshift Data Warehouse, enabling real-time visualization across multiple use cases.
Performed stress testing on data pipelines to ensure scalability and reliability under various load conditions.
Coordinated with cross-functional remote teams for system design and sprint planning.

Candidate Screening Chatbot

Rasa Python NLP PostgreSQL

Led backend development for an AI-powered Candidate Screening Chatbot using the Rasa open-source ML framework, designed to conduct preliminary interview rounds autonomously.
Integrated NLP-driven intent recognition and entity extraction with a PostgreSQL database for candidate data management.
Delivered a fully functional prototype within two months, capable of understanding user messages and generating context-appropriate responses.

University at Buffalo

Dec 2018 – Sep 2019

Data Science Research Assistant

Spatial-Temporal Analysis on GPS Data — NIH Funded

Python R Tableau SSIS MS-SQL

Conducted spatial-temporal analysis on GPS data collected through an NIH-funded survey on travel behavior and social influence.
Built an ETL pipeline to clean, transform, and load data into MS-SQL using SSIS.
Constructed interactive maps in R to visualize user GPS trajectories and cluster most-visited locations.
Analyzed and visualized social influence on travel behavior at both individual and household levels.
Built predictive models using Markov Chain and Conditional Probabilistic Models to forecast user activity by day and time.
Designed a co-location detection program in Python and visualized the resulting social network as a Tableau dashboard.

Niagara Falls Bridge Commission

Sep 2017 – Dec 2017

Data Analyst Intern

Traffic Analysis — Lewiston-Queenston Bridge

Python ML Excel

Performed exploratory data analysis on bridge crossing data to identify traffic patterns and seasonal trends.
Conducted a survey to study traveler behavior and sensitivity to toll cost changes.
Achieved a 13.11% reduction in peak-hour traffic using a Multinomial Logistic Regression model to recommend optimal toll pricing strategies.

WHAT I'VE BUILT

Projects

Python Apache Airflow PostgreSQL FastAPI Next.js Vercel Cloudflare SEC EDGAR API

Featured Project

MktMoves — SEC Filing Intelligence Platform

A production-grade platform I built solo that ingests and surfaces institutional investment intelligence from SEC EDGAR filings for retail investors — the kind of intelligence previously only available through expensive financial data vendors.

Processes 2.6M+ filings across 13F, Form 4, SC 13D, and SC 13G forms dating back to 2015
Architected a 3-stage Airflow pipeline (fetch → XML parse → structured load) with intelligent full/delta load detection and automatic checkpoint resumption on any machine
Designed PostgreSQL schema from scratch including column definitions, indexing strategy, and query optimization for fast UI rendering
Built a rate-limit-aware SEC EDGAR API handler (10 req/sec) and custom XML parser with dedicated error tracking tables for fault tolerance
Designed 10 proprietary analytics metrics: Sector Capital Flow Heatmap, Institutional Crowding Score, Smart Money Divergence Signal, Insider-Institutional Alignment Score, and more
Deployed on lean cost-constrained infrastructure: FastAPI on Railway, Next.js on Vercel, Cloudflare DNS

Python Azure Synapse SQL Apache Spark Power BI Cosmos DB

NYC Taxi Data Analytics — Azure Synapse

Processed three years and 100M+ records of NYC Yellow Taxi data using Azure Synapse dedicated SQL pools and Spark notebooks, then surfaced actionable insights through Power BI.

Configured dedicated SQL pools and Spark pools to manage and analyze 100M+ records across 2021–2023
Ingested, transformed, and loaded 1TB+ of data using Serverless SQL Pool, Spark Pool, and automated Synapse pipelines
Enabled real-time analytics by integrating Synapse Link with Cosmos DB
Identified that 60% of trips were paid by credit card, with Queens showing a unique pattern of higher cash transactions
Built Power BI dashboards revealing Manhattan accounted for 40% of overall taxi demand, with peak demand on Fridays

Python PyTorch Scikit-learn Collaborative Filtering RBM Autoencoder

Movie Recommender System

Built and benchmarked multiple recommendation system approaches including collaborative filtering, deep learning, and matrix factorization to predict user movie preferences.

Developed Item-Based and User-Based Collaborative Filtering models with Cosine, Pearson, and MSD similarity metrics
Trained KNN and SVD++ models to predict user ratings for unseen movies
Achieved 75% accuracy in user preference prediction using a Restricted Boltzmann Machine (RBM)
Enhanced recommendation quality with an Autoencoder model, maintaining an average error of ~1 star rating

Python XGBoost LightGBM Random Forest Feature Engineering Bayesian Optimization

Movie Revenue Prediction

Predicted global movie revenue using advanced ensemble ML models and extensive feature engineering on a real-world dataset with significant missing data challenges.

Performed EDA to remove anomalies, engineered 39 new features, and imputed missing values using predictions from a Bayesian-optimized XGBoost model
Built and compared Linear Regression, Bayesian-optimized Random Forest, XGBoost, and LightGBM models
Achieved an RMSLE score of 0.96

CREDENTIALS

Certifications

Microsoft Certified: Power BI Data Analyst Associate

Microsoft

Certified Tableau Desktop Specialist

Tableau / Salesforce

WHERE I STUDIED

Education

Master of Science in Engineering (Data Science Focus)

University at Buffalo, SUNY

Buffalo, NY · Feb 2019

GPA: 3.73 / 4.0

Bachelor of Engineering in Mechanical Engineering

North Maharashtra University

Jalgaon, India · May 2016

GPA: 8.10 / 10.0

WHAT PEOPLE SAY

Testimonials

“

I was impressed with Sanket's ability to understand and solve problems efficiently. His choice of visualization library was very good, and I was impressed with the way he designed the probabilistic model and Tableau dashboard. Sanket would be a true asset for any positions requiring R, Python, and Tableau.

Dr. Qing He

Professor, University at Buffalo

“

Sanket's creative thinking, expertise, and positive can-do attitude make him an absolute pleasure to work with. He continually delivers results and goes above and beyond. His strengths in staying across issues, pro-actively offering solutions, and being adept at all aspects of communications make him a valuable contributor.

Prachi Sharma

Former Colleague

SanketChavan