Case studies
Production AI, automation, and QA work, written up as engineering documents: problem, architecture, implementation, and results.
Client and employer work in production: AI pipelines, cloud infrastructure, and enterprise QA.
Professional Case Study
28,000+ · Production records
A production AI pipeline that scrapes, enriches, translates, and publishes 28,000+ AI tool listings, with status-gated LLM stages that never skip a failure.
Professional Case Study
34% · Monthly AWS cost reduction
Finding where a production AWS setup spent money it didn't need to, and cutting monthly cost by 34%.
Professional Case Study
20 · End-to-end journeys
Lead QA on a multi-tenant B2B pharmaceutical distribution platform: order-to-cash journeys, business-logic validation, API evidence, authorization, audit-log security, and reconciliation.
ToolPotion production QA and AI education platform QA are summarized under professional experience.
Independent AI systems, written up with the same problem, architecture, and results structure.
AI Engineering Build
Multi-turn · Thread memory across replies
A stateful n8n agent that reads incoming email, remembers the whole thread, answers from verified information, and drafts replies.
AI Engineering Build
~90% · Reported cut in questionnaire turnaround
Collecting live AWS evidence, then answering enterprise security questionnaires with a LangGraph Corrective RAG agent that must cite a verified source for every answer.
Research on making LLM output verifiable, evaluated on an open benchmark.
Research Case Study
83.81% · Reported remediation rate (88/105)
A neuro-symbolic research framework: an LLM proposes Terraform fixes, and a Z3 SMT verifier accepts only patches that provably satisfy cloud security invariants.
Available for remote AI automation, n8n, AI agent, API integration, and B2B SaaS QA projects.