RAG Systems
Retrieval-augmented generation over unstructured documents — chunking, embeddings, and grounded answers with citations back to source.
Arnav Prabhu.
Finance & AI.
Finance and AI. Strategy, risk, and building with models. (And a massive soccer enthusiast.)
Applied AI
Multi-Agent Orchestration
Designing agent hierarchies with specialist roles, task delegation, and real-time cost tracking across model providers.
Agentic Workflows
Coordinating multiple specialized agents — data, analysis, writing, review — into reliable end-to-end pipelines.
Production Delivery
Taking LLM applications from prototype to deployed product with full-stack tooling, evaluation, and iteration.
Selected Work
Axton
AI-powered SEC filing intelligence platform with a RAG pipeline for 10-K, 10-Q, and 8-K filings. Built with Next.js, Supabase, and Gemini AI.
VIEW PLATFORM arrow_outward
pi-swarm
Multi-agent orchestration framework built on pi.dev. Organizes AI agents into a 3-tier company hierarchy with 21 specialist roles. Model-agnostic, real-time cost tracking, and a live terminal UI.
VIEW REPOSITORY code
Doxa
AI-powered equity research pipeline that coordinates six specialized agents across market data, valuation, SEC filings, sentiment, writing, and editorial review to produce high-signal reports.
VIEW REPOSITORY code
“Looking for roles where finance and AI overlap.”
EDUCATION
Arnav Prabhu is a UT Dallas undergraduate pursuing dual B.S. degrees in Finance and Business Analytics & AI. He builds and ships applied AI systems — RAG pipelines, multi-agent frameworks, and LLM applications — with a foundation in finance, risk, and quantitative methods.
EXPERTISE & SKILLS
Experienced in financial analysis, risk management, compliance, and applying AI in finance.
Let's Talk
Open to finance and AI opportunities.