Data Alchemist · Mumbai / Remote

Useful data, not dashboard theatre.

I build ML models, real-time data pipelines, and BI dashboards that move from messy records to decisions — with a taste for systems that look sharp and actually ship.

Hire / Collaborate GitHub LinkedIn
9.73GPA — B.Tech AI & Data Science
10K+events/sec streaming pipeline
540Krecords cleaned for BI analysis
Open to Data Analyst · DS · ML · BI · DE
Location Mumbai or remote roles
Impact 40% faster workflows · 15% yield gains
Portfolio GitHub-first project proof

Work spells.

Resume-backed analytics simulations where the important part was not just charts — it was framing business questions and making recommendations.

BCG X / Forage · Junior Data Scientist

SME customer churn prediction

Scoped churn with BCG’s five-step method, cleaned 30K+ records, engineered price-sensitivity features, and built a Random Forest model with ~75% precision.

30K+ recordsRandom Forest~75% precisionRetention strategy
Tata Group / Forage · Data Analytics

CEO/CMO analytics dashboarding

Cleaned and transformed ~540K rows, validated quantities/prices, calculated revenue, and created executive visuals for seasonality, markets, and top customers.

540K recordsPower BITableauRevenue analysis

Projects with machinery.

Built like systems, not screenshots — forecasting, streaming, automation, monitoring, dashboards.

Analytics Platform

Global Peace Index Analytics Platform

  • Built a Fabric Lakehouse across 23 indicators with 99% uptime and auto-alerts.
  • Forecasted 5+ years using ARIMA and Prophet, improving accuracy by 12%.
  • Delivered Power BI dashboards with animated visuals and custom DAX for risk monitoring.
FabricSparkProphetPower BI
Streaming Pipeline

Real-Time Event Processing Pipeline

  • Designed FastAPI → Kafka → Spark Structured Streaming → Cassandra pipeline.
  • Processed 10,000+ events/sec with 99.9% uptime using partitioning and micro-batch tuning.
  • Automated deployment and monitoring with Docker, Airflow UI, and Kafka Control Center.
FastAPIKafkaSparkCassandra

Capability lab.

Inspired by modern component systems: animated background, orbit visual, radar card, bento proof cards, and command-palette style UI — without the ugly Spline embed.

DATA
Stack

Where I’m strongest

PythonSQLAWSAirflowSparkPower BITableauDockerPostgresFlaskDjangoReactJSGenAIForecastingStorytelling
Education

KJ Somaiya Institute of Technology

Bachelor of Technology in Artificial Intelligence & Data Science, 2022–2026. Current GPA: 9.73.

Profile

Analysis that ships

I combine business framing, clean data work, ML experimentation, dashboard storytelling, and engineering discipline — useful for entry-level analytics, ML, BI, analytics engineering, and data engineering roles.

Need a data person who can make the work feel alive?

I’m looking for Mumbai or remote opportunities where I can build useful analytics products, ML workflows, dashboards, and pipelines.

keyur04mistry@gmail.com+91 91671 19015GitHubLinkedIn