AI Builder · Product · Strategy

I build useful AI products — from strategy to shipped.

I'm Ari Rosenberg — a product leader and builder with 15+ years turning broken processes into systems people rely on. Today I build AI products end-to-end: driving LLM adoption on the world's largest EHR deployment, and shipping my own AI apps from first principles to live.

View projects → Resume (PDF) Get in touch

About

I'm an AI builder, product manager, and strategist. I design and ship AI-native products end-to-end — defining the problem, shaping the product, writing the prompts and the code, and standing up the infrastructure behind it. My instinct is to find what's broken, make the case for change, and then actually build the thing.

For 15+ years at Booz Allen Hamilton I've delivered technology across federal health — from enterprise Power BI ecosystems to generative AI systems. I currently champion AI adoption on the VA's Electronic Health Record Modernization program, the world's largest EHR deployment: building functional LLM and RAG prototypes, authoring the capability roadmap, and navigating enterprise governance toward deployment approval. Along the way I've led product and analytics work across NIH, CMS, and the FDA.

Outside of work I build independently — most recently ACT Journal, a full-stack AI product I took from concept to live. I'm focused on building AI products with genuine utility, and I'm open to roles and collaborations where I can do exactly that: own the problem, shape the product, and ship.

Build

Hands-on across the stack — React, Postgres/Supabase, serverless, and the LLM layer (prompting, evals, caching, multi-provider routing).

Product & Strategy

Frame the problem, define the loop, scope the MVP, and make the hard prioritization calls. Comfortable owning a roadmap and the metrics behind it.

Ship & Iterate

Bias to working software. I get real products in front of real users, instrument them, and improve from evidence.

Product Strategy LLM Apps RAG Architectures Prompt Engineering Evals & Response Eval Python SQL React Supabase / Postgres Power BI / DAX Azure Anthropic / OpenAI / Gemini APIs Stakeholder Alignment Enterprise AI Governance 0→1

Selected Projects

ACT Journal

Featured · AI Product · Solo build

ACT Journal

An AI journaling app grounded in Acceptance & Commitment Therapy. You define your values, write daily entries, and the AI reflects your writing back against what matters — surfacing patterns, tracking behavioral shifts, and closing the loop between insight and action.

Designed and built the full product end-to-end: concept, UX, frontend, database, and the entire AI layer.
Engineered a multi-provider AI proxy (Anthropic, OpenAI, Gemini) with a single normalized streaming interface and server-side key management.
Cut per-user AI cost from ~$2.83/day to ~$0.09/session via prompt caching, three-tier model routing, and context tiering.
Built a persistent "working formulation" so the AI stays calibrated to the person across sessions — not just the latest entry.
Row-Level Security on every table, JWT-authenticated edge functions, and a three-layer entitlement system wired for Stripe billing.

React 19SupabasePostgres + RLSEdge FunctionsClaude / GPT / GeminiPrompt caching

Live site ↗ Source ↗

Enterprise AI · Veterans Affairs (EHRM)

Enterprise LLM & RAG Prototypes

Functional generative-AI prototypes built to transform manual program analysis on the VA's Electronic Health Record Modernization program — the world's largest EHR deployment.

Spotted the opportunity for LLMs to replace painstaking manual lessons-learned analysis, then built working prototypes to prove it.
Developed a RAG-based Q&A system over program documentation and an automated lessons-learned analysis pipeline.
Authored the capability roadmap, secured stakeholder sponsorship, and am navigating VA's enterprise software-approval process toward deployment.

PythonLLMsRAGPrompt EngineeringResponse Eval

Enterprise / government work — details shared on request.

VA EHRM

Pipeline Analytics

Product Leadership · Analytics Platform

Growth & Partnering Analytics Platform

An award-winning, CRM-integrated analytics platform that gave Booz Allen health-market leadership their first-ever real-time view into a multi-billion-dollar growth pipeline.

Identified the lack of pipeline visibility as a strategic gap, then designed and deployed the platform as product lead and developer.
Surfaced the pipeline by technology platform, teaming partners, and competitive landscape — turning scattered CRM data into decisions.
Achieved immediate daily adoption across multiple business units; recognized with an internal award.

Power BIDAXCRM IntegrationData Modeling

Internal Booz Allen project — details shared on request.

Experience & Background

2024 – Present

Independent AI Builder

Self-directed

Design and ship full-stack AI products end-to-end. Most recent: ACT Journal — a values-based AI journaling app built from concept to live, including a multi-provider AI layer, RAG-style retrieval, and cost-optimized infrastructure.

2021 – Present

Analytics & AI Product Development — Veterans Affairs (EHRM)

Booz Allen Hamilton · SAVI Team

Lead analytics and AI product development on the world's largest EHR deployment. Built functional LLM and RAG prototypes, authored the AI capability roadmap, secured stakeholder sponsorship, and replaced fragmented manual reporting with real-time dashboards for lessons learned, deliverables, and acquisitions.

2023 – 2024

Lead Product Developer — Health Market Growth & Partnering Analytics

Booz Allen Hamilton (Concurrent Internal Project)

Designed and deployed an award-winning, CRM-integrated analytics platform giving health-market leadership first-ever real-time visibility into a multi-billion-dollar pipeline. Achieved immediate daily adoption across multiple business units.

2008 – 2021

Product Manager & Analyst — Federal Health

Booz Allen Hamilton · NIH, CMS, FDA

Delivered data and product solutions across federal health agencies: built data-collection applications for a nationwide NIH longitudinal study, analyzed $2B+ in ACA grant funds at CMS, and reengineered evaluation of 100k+ consumer comments — consistently finding what was broken and shipping the fix.

Education

B.S., Finance & Information Systems

University of Maryland, College Park

Minor in Rhetoric · Full-Tuition Scholarship

Certifications

PMP · Lean Six Sigma (Yellow Belt) · ITIL Foundation

Let's talk.

Open to roles, collaborations, and conversations about building with AI. The fastest way to reach me is email.

rosenberg.ari@gmail.com LinkedIn GitHub