Hrushikesh Thanikonda
AI/ML Engineer · Data Scientist · Jersey City, NJ (NYC metro)
AI/ML engineer and data scientist with an MS in Data Science from Stevens Institute of Technology (May 2026). My graduate work spans financial time-series analysis, ETL pipeline development, and Azure-based data platforms. I build reliable, auditable data systems — pipelines with built-in quality checks, statistically validated forecasting models, and governed data access — using Python, SQL, and Azure.
Education
Stevens Institute of Technology — Hoboken, NJ
Master of Science, Data Science · Sept 2024 – May 2026
Blekinge Institute of Technology — Karlskrona, Sweden
Bachelor of Science, Computer Science and Engineering · May 2021 – Aug 2023
Project Experience
AI-Assisted Portfolio Engineering — this site
June 2026 – July 2026
- Built a production Next.js/TypeScript portfolio spec-first with an AI coding agent under human review gates: 19 approved specification documents (PRD, IA, design system, performance budget) preceded implementation.
- Enforced a strict-TypeScript, single-typed-content-layer architecture with server-rendered content, semantic HTML, and reduced-motion accessibility paths.
Forecasting & Market Trend Analysis
Sept 2025 – Dec 2025
- Modeled BlackRock equity price series with ARMA models, selecting orders through ACF/PACF diagnostics rather than automated search.
- Modeled 50 years of macroeconomic data (California unemployment, 1976–2025) with SARIMA to capture seasonal and cyclical structure.
- Gated all forecasts behind holdout testing, confidence intervals, and residual diagnostics before decision-support use.
Cloud-Based Data Ingestion & Storage Platform
Sept 2025 – Dec 2025
- Designed and built an Azure data platform (.NET, Azure SQL, Cosmos DB, Azure Storage) with three-role access control and human-in-the-loop approval gating all downstream data access.
- Separated storage layers by data shape — constrained relational metadata in Azure SQL, image binaries in Cosmos DB — keeping schemas enforceable.
- Eliminated hard-coded credentials via Azure Key Vault and tracked all access events for audit.
- Implemented ingestion-path validation and failure detection.
Experimental Data Pipeline — Action Generation (EPIC-Kitchens)
Sept 2025 – Dec 2025
- Engineered a video-to-image ML data pipeline: timestamp-aligned frame extraction, stacked (t, t+Δt) input pairs, MediaPipe noise filtering, and augmentation to mitigate dataset bias.
Insurance Data Engineering & Reporting Pipeline
Jan 2025 – May 2025
- Built Python/SQL ETL pipelines with schema-first design and explicit validation rules.
- Implemented data-quality checks — null handling, outlier detection, distribution-shift monitoring — to catch pipeline issues early.
- Produced validated datasets consumed by reporting and modeling layers; built Power BI dashboards surfacing trends and anomalies for stakeholder decisions.
Performance of Different ML Models — Cyberbullying Dataset
Sept 2024 – Dec 2024
- Annotated and iteratively re-reviewed 40,000+ text samples with human-in-the-loop cycles, improving label consistency; preprocessing contributed to 84% classification accuracy.
Undergraduate Research — Sentiment Analysis of Cyberbullying Tweets
May 2021 – Sept 2023
- Developed and published an SVM-based sentiment classifier (84% accuracy, 40,000+ samples) as a bachelor's research thesis.
View publication
Technical Skills
Data Engineering & Pipelines
- Python ·
- SQL ·
- PostgreSQL ·
- ETL/ELT workflow design ·
- data ingestion & transformation ·
- schema design ·
- data validation ·
- data quality checks (nulls, outliers, distribution shift) ·
- failure detection
Machine Learning & Forecasting
- Time-series modeling (ARMA, SARIMA) ·
- ACF/PACF model diagnostics ·
- forecast validation (holdout, confidence intervals, residuals) ·
- classification (SVM) ·
- data annotation & human-in-the-loop labeling ·
- data augmentation
Cloud & Platform (Azure)
- Azure SQL ·
- Cosmos DB ·
- Azure Storage ·
- Azure Key Vault ·
- role-based access control ·
- .NET backend services ·
- secrets management
Analytics & Reporting
- Power BI dashboards ·
- statistical diagnostics ·
- anomaly surfacing ·
- decision-support analytics
Engineering Practices
- Git & GitHub ·
- AI-assisted development (spec-driven, review-gated) ·
- TypeScript & Next.js ·
- pipeline reliability ·
- documentation & reproducibility ·
- human-in-the-loop review workflows
Certifications
- Microsoft Power BI Data Analyst Professional Certificate
- Meta Database Engineer Professional Certificate