System Online - All Pipelines Active

I don't just move data.

I engineer intelligence.

Data Engineer & AI/ML Specialist with 5+ years building enterprise-scale pipelines and machine learning systems. Currently building anomaly detection across 1M+ healthcare providers for government program integrity, after saving $1.6M+ annually with AI at American Express.

|
Resume
0M+
Providers Analyzed
+0%
ML Detection Lift
$0M+
Annual Savings
0%
Pipeline Optimization

// ABOUT

The Engineer Behind the Systems

about_tanish.log

> INIT: Career trajectory loaded (newest first)...
> STATUS: Evolving from Data Engineer → AI/ML Engineer
> MISSION: Building systems that think, not just process.
[2026] ML anomaly detection - 1M+ Medicaid providers, govt program integrity
[2026] Current: Associate Data Engineer-AI at Jorie AI
[2025] Led multi-cloud migration (AWS → GCP)
[2024] Optimized ETL pipelines - cut processing from 8hrs to 3.2hrs
[2023] Built AI data scrubbing system - saved $1.6M annually
[2021] Joined American Express - Arena Testing Team
[2020] First data role at CE-CERT - processing 500GB+ vehicle telemetry
[2020] MS Computer Engineering - UC Riverside
[2019] Graduated BS Computer Engineering - UC Riverside

I started my journey at UC Riverside, where I earned both my BS and MS in Computer Engineering. What began as curiosity about how systems process information evolved into a passion for building intelligent data architectures at scale.

At American Express, I spent nearly 5 years on the Arena Testing Team, where I didn't just build pipelines - I engineered solutions that fundamentally changed how the team operated. My AI-based data scrubbing system eliminated $1.6M in annual vendor costs. My ETL optimizations cut processing time by 60%. My real-time streaming pipelines validated 130K+ daily test transactions at sub-second speed.

Now at Jorie AI, I'm applying that same systems-thinking across both sides of the business - RCM dashboards that turn complex billing data into actionable intelligence, and ML anomaly detection across 1M+ Medicaid providers for government program integrity. That project is where the shift became real: engineered features, unsupervised and supervised models working together, and a 39% detection lift over rules alone. My trajectory is clear: from moving data to engineering intelligence.

5+

Enterprise Experience

MS

UC Riverside

3

Domains

// EXPERIENCE

Systems I've Built & Scaled

Jorie AI

Associate Data Engineer - AI

Feb 2026 - PresentOak Brook, IL

CURRENT
>

39% detection lift over rules alone - ML anomaly detection across 1M+ Medicaid providers for government program integrity review

>

Engineered 40 provider-level features with Isolation Forest, Autoencoder & LightGBM under a positive-unlabeled framework

>

Designing interactive RCM dashboards tracking AR aging, denial rates, and reimbursement trends

>

Building secure SQL Server pipelines for healthcare billing analytics

>

Implementing automated data validation for billing and claims accuracy

>

Translating complex RCM requirements into scalable reporting solutions

PythonLightGBMPyTorchPolarsSHAPSQL ServerPower BIHealthcare Analytics

American Express

Software Developer / Data Engineer

Apr 2021 - Feb 2026Phoenix, AZ

4.8 YEARS
>

$1.6M saved annually - AI-based data scrubbing for banking personas

>

60% faster - ETL optimization with parallel processing using PySpark & AWS Glue

>

99.7% data accuracy - Automated 15+ manual workflows with real-time validation

>

130K+ daily test transactions - Real-time streaming test transaction validation via AWS Lambda & Kinesis

>

45% improvement AWS → GCP - Query performance boost through multi-cloud migration

>

67% fewer incidents - Comprehensive CI/CD implementation for data pipelines

PythonPySparkAWSGCPBigQueryLambdaKinesisHDFSPower BISSIS

CE-CERT, UC Riverside

Software Engineer Intern / Data Analyst

Jun 2020 - Jan 2021Riverside, CA

INTERNSHIP
>

Built automated data collection for Eco-Vissim driving simulation

>

500GB+ - Vehicle telemetry data processing for fuel efficiency research

PythonTableauETLSensor Data
// PROJECTS

Intelligence Systems Portfolio

Real-world systems engineered to detect, predict, and optimize at enterprise scale.

39% Detection Lift

Medicaid Provider Anomaly Detection

End-to-end detection system built at Jorie AI for government program integrity, analyzing 1M+ Medicaid providers across public CMS utilization, NPPES registry, and OIG LEIE datasets. Rule-based peer benchmarking flags providers billing 3x+ their state/specialty norms; an ML layer with 40 engineered features adds unsupervised (Isolation Forest, Autoencoder) and supervised (LightGBM) detection under a positive-unlabeled framework with leakage guards, plus SHAP explainability and executive dashboards for non-technical reviewers.

1M+Providers Analyzed
32,714Found by ML Only
17,737High-Priority Overlap
Python
LightGBM
PyTorch
Polars
SHAP
Scikit-learn
$1.6M Annual Savings

AI Data Scrubbing Engine

Developed an AI-powered data scrubbing solution for banking and credit card personas, completely eliminating dependency on expensive external vendor contracts for test data generation.

$1.6M/yrCost Eliminated
0Vendor Dependencies
99.7%Data Quality
Python
Machine Learning
PySpark
AWS
Current Work @ Jorie AI

Healthcare RCM Analytics

Building interactive Revenue Cycle Management dashboards that track AR aging, denial rates, reimbursement trends, and collection performance - backed by secure SQL Server pipelines and automated validation checks that verify billing and claims accuracy before publication.

AR · Denials · CollectionsKPIs Tracked
AutomatedData Validation
Healthcare RCMDomain
SQL Server
Power BI
Python
Healthcare Data
Sub-second Validation

Real-time Streaming Pipeline

Designed and deployed real-time streaming data pipelines using AWS Lambda and Kinesis for test environment validation, enabling sub-second processing of 130K+ daily test transactions at scale.

<1 secLatency
130K+ txnsDaily Test Volume
99.9%Uptime
AWS Lambda
Kinesis
Real-time Streaming
Event-driven
100+ Engineers Supported

Enterprise Data Lake

Architected enterprise-scale test data lake on AWS, consolidating data from Oracle, Teradata, mainframes, and RDBMS sources into HDFS, supporting massive QA operations.

4+ typesData Sources
100+Engineers Supported
5TB+Data Migrated
AWS S3
HDFS
Oracle
Teradata
Hadoop
60% Faster Processing

ETL Performance Optimizer

Optimized ETL pipeline performance through parallel processing and incremental loading strategies, cutting processing time from 8 hours to 3.2 hours.

60%Speed Improvement
8 hoursBefore
3.2 hoursAfter
PySpark
AWS Glue
Parallel Processing
SSIS
20% Cost Reduction

Multi-Cloud Migration

Led migration from AWS to GCP for test data infrastructure, achieving significant cost reduction while improving query performance through BigQuery optimization.

20%Cost Savings
+45%Query Performance
5TB+Data Migrated
AWS
GCP
BigQuery
Data Migration

// SKILLS

Technology Arsenal

AI / ML & Data Science

Python / PySpark95%
Anomaly Detection88%
Probabilistic Modeling85%
Gradient Boosting (LightGBM)85%
Deep Learning (PyTorch)78%
Model Explainability (SHAP)80%
Predictive Analytics80%

Cloud & Big Data

AWS (S3, Glue, Lambda)92%
GCP (BigQuery, Dataflow)85%
Hadoop / Hive / Spark88%
Real-time Streaming82%
Data Lake Architecture90%
Kinesis / Kafka78%

Engineering & Tools

ETL / Data Pipelines95%
C# / .NET / Scala75%
SQL / NoSQL90%
Power BI / Tableau85%
CI/CD Automation82%
SSIS / Automation80%

Tools & Platforms

Python
PySpark
Scikit-learn
LightGBM
PyTorch
SHAP
Polars
Pandas
Plotly
Scala
C#
.NET
SQL
AWS S3
AWS Glue
AWS Lambda
AWS Kinesis
GCP BigQuery
GCP Dataflow
Hadoop
HDFS
Hive
Spark
Oracle
Teradata
SSIS
Power BI
Tableau
Git
CI/CD
Docker
Terraform

// INITIATE CONNECTION

Let's Build Something Intelligent Together

Whether you're looking for a Data Engineer who thinks in systems, an AI/ML specialist who delivers measurable impact, or a collaborator who turns complex data into strategic advantage - let's connect.

> tanish.arora

Engineered with precision. Powered by data.

© 2026 Tanish Arora