AI application builds
- LLM assistants
- Retrieval-augmented generation
- OpenAI API
- Anthropic Claude API
- Context strategy
- Guardrail design
- Workflow automation
AI Engineer & Builder
I build and ship AI products end-to-end, and I'm comfortable doing it close to the customer — turning a vague business problem into a deployed, working system. Full-stack across Next.js and Python/FastAPI, with a delivery background behind it.
Based in Johannesburg. The thread across my builds: AI is commentary, never the source of truth — the load-bearing logic stays deterministic. Behind the engineering is years of client-facing delivery: scoping, stakeholder alignment, QA, launch, and handover.

My background spans software development, digital strategy, reporting, performance thinking, and enterprise project leadership. That range lets me move between technical detail and operational reality without losing the thread.
I treat AI initiatives as engineered systems — with reliability standards, governance, human oversight, and measurable adoption inside the business. The interesting work is rarely the model itself; it's the surrounding architecture that makes the model dependable.
Building AI agents that close the loop on a real task — combining LLM reasoning with tool use, retrieval, and execution behind a single surface.
Practical assistants and decision-support tools with grounded context, retrieval where it matters, and outputs the user can trust.
Connecting AI into existing business workflows so they run faster, more consistently, and with less manual handling.
Leading full project lifecycles, stakeholder alignment, and handovers that keep strategy tied to execution.
Using analytics, reporting, and KPI frameworks to measure impact and continuously improve outcomes.
Hands-on background in software and web development with strong full-stack and platform fundamentals.
Hands-on development skills built across software delivery, WordPress engineering, AI tooling, and production web projects.
Leading end-to-end delivery, stakeholder governance, and operational handovers across complex initiatives.
Drove digital transitions with precise scheduling, budget control, and cross-team alignment.
Built reporting clarity and trained teams during new system rollouts.
Facilitated technical workshops and strengthened hiring requirements for web-focused roles.
Delivered digital growth strategies through freelance consulting and omni-channel campaign execution.
Led channel strategy, social performance, and data-informed campaign execution across multiple organizations.
Managed full-stack project delivery, team leadership, and quality assurance under demanding timelines.
Built custom websites, plugins, and client-ready digital platforms with strong UX and performance focus.
Ask about the work, the decisions, the skills, and the way I think.
Live products and working prototypes across AI, legal tooling, property, specialist services, and medical support — each opens a short problem-to-result write-up alongside the running app.
Built autonomous agents, RAG-powered voice agents, and a multi-agent go-to-market system. Covered LLM integration (OpenAI, Gemini, Anthropic), Supabase vector RAG, MCP, webhooks, and JSON data pipelines.
Shipped live products with agentic coding workflows — including this site and its digital twin, and the PreLegal assistant. Covered Claude Code, MCP, skills, sub-agents, multi-agent orchestration, and the Claude Agent SDK.
Six-week intensive building autonomous agents across every major framework — OpenAI Agents SDK, CrewAI, LangGraph, and AutoGen — starting from raw LLM design patterns and culminating in MCP (Model Context Protocol). Eight hands-on projects, capped by a trading-floor capstone: four agents making autonomous trades over six MCP servers and 44 tools.
Three-module workshop on leading AI initiatives with measurable commercial impact: strategy and investment frameworks, the scaling laws and cost/benefit of techniques from RAG to reasoning, the risks and opportunities of agentic AI, and AI architecture, org design, governance, and roadmap decisions from a commercial perspective.
Four completed tracks of an applied AI engineering curriculum. The FinAlly and PreLegal portfolio projects above were built as part of it.
Academic program co-badged with University College Cork, Ringling College, and Ireland's Higher Education Authority & National Forum.
Agentic coding with Claude Code — the workflow behind the builds in this portfolio.
Building on the Claude developer platform and API.
Core foundations of working effectively with Claude.
Delegating multi-step work to Claude as an autonomous coworker.
Applying AI to real small-business workflows and operations.
Based in Johannesburg, open to forward-deployed and AI engineering roles, plus build engagements across applied AI products, automation workflows, and delivery-led programs.