AI Strategy · Applied AI · Decision Systems
USC Marshall MBA '28 · Former ML & Conversational AI Engineer
My background is in machine learning and conversational AI. My current work sits one layer above the model itself:
Where should AI be deployed? What evidence should a decision rely on? How do you make assumptions explicit? And how do you turn messy information into something a decision-maker can actually use?
I build analytical products at the intersection of technology, strategy, finance, and decision science — usually around questions where the answer is less interesting than the reasoning that produces it.
Where should a company expand next? A decision-support system for comparing US metropolitan markets using public Census data, explicit strategic assumptions, adjustable weighting, and Monte Carlo sensitivity analysis. Decision problem: Market selection |
What can public evidence tell us about the value of a music catalog? An underwriting workbench that reconstructs public catalog evidence, makes economic assumptions explicit, models cash flows, and tests valuation across scenarios and sensitivities. Decision problem: Asset underwriting |
Does management allocate capital in line with the strategy it communicates? A research tool that extracts strategic claims from SEC filings and tests them against structured financial evidence. Decision problem: Strategy credibility |
More tools that turn ambiguous strategic questions into explicit decision models. Areas I'm especially interested in: AI transformation |
| Strategy | AI & Engineering | Decision Science |
|---|---|---|
| Market entry | LLM applications | Scenario analysis |
| AI transformation | Agents & RAG | Sensitivity analysis |
| Business models | Data pipelines | Explicit assumptions |
| Capital allocation | Applied ML | Evidence quality |
| Competitive analysis | Product engineering | Valuation |
The technical system matters. So does the business question it is being built to answer.
Before business school, I worked in machine learning and conversational AI, building systems across NLP, generative AI, and enterprise knowledge applications.
I studied Computer Science at BITS Pilani, Dubai, where my academic work included research in Natural Language Generation and generative AI before the current LLM wave.
I'm now pursuing my MBA at USC Marshall, focusing on the intersection between technical feasibility and strategic value: how emerging technology changes business models, operating models, capital allocation, and competitive advantage.
Earlier AI / ML work
My older repositories document how I got here — NLP systems, conversational AI, recommendation systems, ML experiments, and generative-language projects built during university.
A few of them:
- MedBot — generative conversational system built on distilled GPT-2 for medical-domain dialogue
- Villager — open-domain conversational generation using GPT-2 and Topical-Chat
- Luna — NLP-based conversational assistant
- Heart Disease Prediction — comparative ML research project
- Collatz Conjecture — procedural generative-art experiment
I keep these repositories public because they show the technical foundation underneath the work I'm doing today.