syed mutaib ali

SAT Knowledge Matching System

A hybrid retrieval engine that maps student questions to relevant SAT curriculum knowledge points in real time. Built with a custom two-stage pipeline combining TF-IDF keyword retrieval (built from scratch, no sklearn) with a semantic heuristic booster that handles domain alignment, tag overlap, and key concept matching. Covers 30 SAT knowledge points across Math and Reading & Writing. Deployable as a Streamlit web app or CLI tool, with a clean architecture designed to drop in sentence-transformer embeddings as a future upgrade.

Spotify ETL Pipeline on AWS

End-to-end serverless ETL pipeline that extracts playlist and track data from the Spotify API, transforms it into structured datasets, and loads it into Amazon S3 for analytics. Extraction is triggered daily via CloudWatch, with a second Lambda function auto-triggered on S3 upload to transform raw JSON into normalized CSV tables (albums, artists, songs). Schema is inferred automatically via AWS Glue Crawler and exposed for SQL querying through Amazon Athena.

Content-Based Book Recommendation System

Recommendation engine that suggests similar books based on user rating patterns and book metadata. Built a user-item matrix from 250k+ ratings across three datasets, applied cosine similarity to surface the top matches for any given title, and filtered for statistical reliability by requiring a minimum of 250 ratings per book. Outputs are serialized via pickle for deployment in a web interface.

Data Scientist

State Street

01/01/2025 - Present

• Designed predictive liquidity risk models using XGBoost, LSTM time-series networks, and Bloomberg B-PIPE, forecasting portfolio liquidity stress up to 30 days ahead, reducing liquidity risk incidents by 32% across multi-asset portfolios. • Implemented real-time anomaly detection pipelines using Apache Kafka, AWS MSK, Isolation Forests, and Autoencoders, processing 10M+ market events daily, cutting risk detection latency by 45% with sub-200ms response times. • Evaluated and deployed explainable AI frameworks using SHAP, LIME, and Snowflake SQL, providing transparent risk drivers, successfully passing SEC and internal audits with zero model governance findings. • Architected end-to-end MLOps workflows using MLflow, Kubeflow, Docker, Kubernetes on AWS EKS, automating model deployment and monitoring, improving forecasting accuracy by 28% and deployment reliability by 40%. • Developed self-service analytics dashboards using Tableau, Power BI, Plotly Dash, and FIS/SimCorp integrations, enabling portfolio managers to identify stress scenarios 2–4 weeks earlier.

Data Scientist

Persistent Systems

01/05/2021 - 01/07/2023

• Engineered ML-based credit scoring models using Logistic Regression, Random Forest, Gradient Boosting, and CIBIL/Experian/Equifax APIs, processing 1M+ loan applications per month, reducing loan default rates by 27% and improving approval speed by 60%. • Developed advanced behavioral feature engineering pipelines using Python, Pandas, PySpark, UPI transaction analytics, and Account Aggregator (AA) Framework, increasing approval rates for new-to-credit customers by 22%. • Implemented explainable AI models using SHAP, StatsModels, and Python, ensuring 100% RBI compliance for model transparency and credit risk reporting. • Designed low-latency fraud detection pipelines using Isolation Forest, Autoencoders, Apache Kafka, Spark Streaming, and NPCI/Payment Gateway APIs, detecting 92% of fraudulent transactions in milliseconds while reducing false positives by 35%. • Orchestrated cloud-native ML workflows using AWS S3, EC2, Lambda, Docker, Kubernetes, MLflow, and Jenkins, reducing end-to-end model deployment time by 50% and supporting 20M+ transactions/day.

Master of Science, Information Technology and Project Management

Arizona State University, USA

01/08/2023 - 01/05/2025

Bachelor of Engineering, Computer Science

Visvesvaraya Technological University, India

01/08/2018 - 01/05/2022

AWS Academy Cloud Architecting

Data Analyst Certification