Independent Consulting · GCP + AI Systems
I architect cloud infrastructure and build AI/LLM systems that hold up beyond the demo. GCP landing zones, RAG pipelines, agent orchestration, and PII-safe LLM gateways — engineered to be private, observable, and boring in the right ways.
STACK
GCP · Anthropic · OpenAI
SPECIALTY
RAG · Agents · PII
SHIPPED
iOS · Android · Web
AVAILABILITY
Q4 2026
§ 01 / Expertise
Founders, CTOs, and platform teams who need AI systems to work in production — not slide decks.
Landing zones, org policies, IAM boundaries, VPC design and cost + compliance guardrails for regulated workloads. Certified GCP practitioner shipping production infra since 2023.
End-to-end LLM systems: pgvector RAG, agent orchestration, prompt engineering, evaluation harnesses, and PII-redacting gateways in front of Anthropic / OpenAI / local models.
Solo-shipped a consumer AI app from architecture to Apple + Google Play release. I bring the same discipline to client work: no half-built demos, only systems that survive users.
§ 02 / Case Study
A consumer AI app I designed, built, and shipped end-to-end. Students drop in notes, PDFs or lecture recordings; the app returns structured flashcards with active recall built in.
Problem
Students spend hours turning notes into study material — manually creating flashcards, re-typing PDFs, transcribing lectures. Existing tools break on real-world source material and force manual card creation.
Approach
A mobile-first pipeline that ingests notes, PDFs and lecture material, extracts concept-level structure with an LLM prompt pipeline, and generates flashcards ready for spaced-repetition review. Onboarding designed for a 60-second first flashcard.
Outcome
iOS shipped and live on the App Store. Android completed closed testing and moved into production. Testing cycle drove ASO revamp, onboarding walkthrough and UX hardening. Targeting 10k–100k first-year installs.
§ 03 / Work with me
30-minute discovery call over Google Meet, no pitch. I'll tell you if I'm the right fit — or who is.