Projects

Each project framed as a case study: problem, discovery, solution, outcomes.

Store-intelligence dashboard: task completion, on-shelf availability, response distribution, and per-department performance

Multi-Tenant Retail Analytics Platform

One config-driven codebase I wrote and maintained, serving six enterprise grocery and convenience retailers across four countries in three currencies

6 orgs / 4 countriesone config-driven analytics codebase I wrote and maintained, in three currencies

Retail AIPlatform ArchitectureAnalytics MethodologyMetric Governance

PythonSQLite / PostgresPlaywrightn8nChart.js

Read case study

AI Insight Engine: Deterministic Where It Matters

A production LLM pipeline for retail execution data where the model never computes a dollar figure

2,155 → 1found the schema constraint that forced the model to invent store counts, then fixed it

AI ArchitectureLLM EvaluationPrompt GovernanceProduct Design

PythonClaude (Opus judge)FastAPISQLiteYAML policy layer

Read case study

Architecture of the test engine: an AI lane that authors scenarios and triages evidence, and a deterministic lane that runs specs through fixtures, calibration, device runs, a reference-model oracle, and an evidence package

Evidence-Audited Test Engine for an AI Retail App

The AI authors and triages; deterministic code delivers every verdict

8 weeksfrom no automated regression capability to gated runs on real devices

Test ArchitectureAI-Native EngineeringEvidence Discipline

PythoniOS device automationJSON Schemapytest

Read case study

Catalyst CPQ marketing site — live at catalystcpq.io

B2B CPQ Sales SaaS

Multi-tenant configure-price-quote SaaS for industrial distribution — my own venture, built end-to-end

Full stackBuilt end-to-end and deployed; first prospect onboarded with their 711-SKU catalog

0-to-1 Product DevelopmentTechnical PMB2B SaaS

Next.js 14TypeScriptFirebaseTailwind

Read case study

Detailed multi-agent architecture diagram: hub-and-spoke knowledge sync, typed agent roles, RAG pipeline, and hook enforcement

AI Operations & Orchestration Ecosystem

The system I actually work inside every day, where governance is enforced by hooks that can refuse a tool call

Refusesthe enforcement layer blocks unverified commits and undeliverable agents at the tool layer, daily

System DesignAI OperationsAutomationDeveloper Experience

PythonTypeScriptLlamaIndexFastAPIn8nDocker

Read case study

AGC Cost & Performance view: API usage, subscriptions, per-session cost, daily spend by provider, cost by agent role

Agent Governance Console

One place to see what the agents did, what they decided, and whether the practice behind these builds is healthy

One surfaceaudit trail, decisions, and health scoring for the agent practice behind the builds on this site

Governance DesignObservabilityFull-Stack Development

Next.jsTypeScriptPostgreSQLElectron

Read case study