CLIMBING THE TOWER...
WEB FIRED 000%
SCROLL 000%
// AGENTIC AI . EXPLAINABLE ML . FULL-STACK AI - HYDERABAD, INDIA

I BUILD AI SYSTEMS. THAT SHIP.

AI & Data Science student building meaningful AI systems, agentic workflows, and full-stack ML products that move from idea to deployment.

Meaningful AIAgentic WorkflowsExplainable MLFull-stack AIML Governance Meaningful AIAgentic WorkflowsExplainable MLFull-stack AIML Governance
Build.Explain.Ship.Measure.Improve.Build.Explain.Ship.Measure.Improve. Build.Explain.Ship.Measure.Improve.Build.Explain.Ship.Measure.Improve.
01 - Build Style

I LIKE BUILDING SYSTEMS THAT ARE MEANINGFUL

I am drawn to AI products where the model is only one part of the system. The best builds have a clear problem, a reliable pipeline, a usable interface, and explanations people can trust.

My projects sit around agentic AI, explainable ML, ML governance, and full-stack product engineering - the kind of work that survives beyond a demo screen.

"Useful beyond demos is the bar."

01
Meaningful
02
Explainable
03
Fast
04
Deployable
05
Useful Beyond Demos
02 - Featured Builds

CASE FILES FROM THE BUILD LAB

CASE #01

AlterScore

Alternative credit scoring for unbanked borrowers, turning behavioral, cognitive, and text signals into a 300-850 score with explainable recommendations.

XGBoostSHAPFastAPIReact
CASE #02
🔍

Pathfinder

An AI-assisted attack-path finder that models networks as weighted graphs, uses A* search, and ranks patch strategies by mitigation impact.

A* SearchNetworkXFastAPIDocker
LAB NOTES
🤖

Agentic Systems

Exploring agentic workflows, harness engineering, evaluation loops, and product patterns that make AI systems more reliable and useful.

AgentsHarnessesEval LoopsML Products
03 - Now Building Around

THE CURRENT SIGNAL STACK

01

Agentic AI

Workflows where tools, memory, retrieval, and decisions are designed as a system, not just prompted into existence.

02

Harness Engineering

Building the scaffolding around models: evaluations, traces, checks, fallbacks, and controlled execution paths.

03

Explainable ML

Models that can show why they made a decision, what changed the output, and where the system is uncertain.

04

ML Governance

Drift tracking, fairness checks, counterfactual stability, and the boring-but-important pieces that make ML safer.

05

Full-stack AI Products

End-to-end products with usable interfaces, APIs, deployments, and feedback loops - not just notebooks.

04 - Contact

BUILD SOMETHING MEANINGFUL?

Reach out for projects, internships, collaborations, or a good conversation about AI systems that should actually work.