// THE HUMAN BEHIND THE PIPELINE

I MAKE DATA BEHAVE.
AND AI DOES THE HEAVY LIFTING.

Engineering graduate. Big Four Data Engineer. AI-powered builder. Currently turning “what if?” into “it actually works.”

IDENTITYHUMAN (MOSTLY)
OCCUPATIONDATA ENGINEER
SIDE QUESTBUILDING WITH AI
CURRENT MODEEXPERIMENTING
BUGS DETECTEDCLASSIFIED
[ EXPLORE MY UNIVERSE ][ INITIATE CONTACT ]
Type ./discover_human.sh (or help) and press Enter.

// SYSTEM DIAGNOSTIC

The short version. The slightly unhinged version.

Started with engineering. Entered the world of enterprise data. Learned that moving data between systems is somehow both a science and a full-time detective job. Then discovered AI could turn ideas into actual software. Now I build, experiment, automate, and occasionally wonder if I should have just opened a chai shop.

// CLASS: DATA ENGINEER / AI BUILDER

Character select: the data-powered builder

Real skills first, jokes clearly labelled. The meters are decorative: no fake proficiency scores here.

ABILITIES JOKES

PIPELINE DETECTIVE (passive): can spend an unreasonable amount of time discovering why one record went missing.
AI COPILOT (special): turns rough product ideas into prototypes faster.
NULL SURVIVOR (ultimate): faces unexpected NULL values and questionable source data.
DEBUGGING INSTINCT (cooldown): unknown. Depends on how badly production is behaving.

// THE EXPERIMENT LAB

Things I built instead of just thinking about building.

Statuses are honest: a prototype is labelled a prototype.

Browse all public repositories ↗

// FIELD NOTES

Lessons from building in public.

Practical notes on data, AI experiments, and the small decisions that make software easier to trust.

Read the blog →

// 02:07 AM

My brain doesn’t have a sleep mode. It has a backlog.

Click a thought. Then click the one you probably shouldn’t.

// SCRIPTED, NOT AI (YET)

Ask the digital intern

A local scripted bot with predefined answers, no model connected. It’s structured so a real API can be swapped in later through a server-side endpoint (keys never in the browser).

// CORPORATE MODE ACTIVATED

Yes, I can be professional. There’s proof.

A few highlights from my recent work: reliable data delivery, careful reconciliation, and practical deployment governance.

—

· Pune, India (Remote)

    Stack:
    Awards & extras:

    “I can discuss enterprise data workflows in a meeting and then spend the evening trying to build an app that solves a problem three people on the internet have.”

    // DEPLOY A CONVERSATION

    Enough scrolling. Let’s build something.

    Got an interesting idea, a difficult data problem, a potential opportunity, or a product that sounds slightly ridiculous but might actually work? I’m listening.

    COLLABORATION STATUS: OPEN TO INTERESTING IDEAS

    Email meConnect on LinkedInExplore my GitHub

    Or write to