β¨ Transitioning from web development into data analytics, I fuse a builder's instinct with an analyst's curiosity to turn raw numbers into decisions people actually act on.
I live at the intersection of:
π Data & Business Intelligence
π» Software Engineering
π€ Generative AI
π¨ Creative Web Experiences
I bring a strong commercial lens to everything I build β dashboards aren't just charts, and code isn't just syntax. Both need to move the needle. π
flowchart TD
A["π RAW DATA"] --> B["π‘ INSIGHT"]
B --> C["πΌ BUSINESS VALUE"]
B --> D["π PRODUCT VALUE"]
C --> E["π IMPACT"]
D --> E
style A fill:#0D1117,stroke:#00F0FF,stroke-width:2px,color:#00F0FF
style B fill:#0D1117,stroke:#7B2CBF,stroke-width:2px,color:#7B2CBF
style C fill:#0D1117,stroke:#0066FF,stroke-width:2px,color:#0066FF
style D fill:#0D1117,stroke:#FF0080,stroke-width:2px,color:#FF0080
style E fill:#0D1117,stroke:#FFD700,stroke-width:2px,color:#FFD700
π― Core Focus: Extracting intelligence π§ , engineering seamless applications π, and weaving Generative AI π€ into everyday workflows.
| Area | What I'm doing |
|---|---|
| πΒ Β Data Analytics | Turning messy data β meaningful insights |
| π€Β Β Generative AI | Exploring AI-powered workflows & products |
| πΒ Β Full Stack | Building useful things for the web |
| π§ͺΒ Β Experimenting | Always breaking things to understand them |
π² Click for a random dev-fact about me
π I do my best debugging after midnight. β Coffee-to-commit ratio: roughly 1:1. π I've stared at more dashboards than Netflix recommendations. π§© Half my browser tabs are Stack Overflow. The other half is documentation I'll "read later."

