AI & Automation Architect · LLM Specialist · AI Engineer
AdelusiDan Kunle.
Shipping production AI products, agent systems and automation pipelines for clients across UK · US · Canada · Dubai · and beyond.
Orchestrating AI across the full stack

By the numbers
Selected work
Products built. Clients served. Multiple countries.
Services
Five lanes. One operator.
Production AI products with Claude, Gemini, OpenAI, DeepSeek, OpenRouter — multi-agent orchestration, RAG, tool use, and reliable output.
Deliverables
- →LLM-powered products shipped end-to-end
- →Multi-agent orchestration and function calling
- →RAG with source citations
- →Prompt engineering and evaluation pipelines
- →Cost + latency optimisation
GeminiClaudeOpenAIOpenRouterDeepSeekVector DBsNext.js
Engagement types
Pick the shape that fits your org.
Freelance sprints, long-term contracts, remote full-time, advisory, or AI training — I take work in all these shapes.
01
Freelance
Fixed-scope builds, 2–12 weeks. Ideal for MVPs, prototypes, and well-defined features.
02
Contract / Retainer
Ongoing engagement, 3–12 months. For teams shipping AI features continuously.
03
Remote Full-Time
Founding engineer, senior AI engineer, or LLM specialist roles with global teams.
04
Advisory
Architecture review, AI roadmap, due diligence — for founders and technical leadership.
05
AI Training & Eval
RLHF, preference ranking, code correctness, factuality — Outlier, Mercor, Scale AI work.
Experience
The work so far.
2024-01 — Present
Remote-first
Founding Member & Lead AI Specialist
Mabi Labs
- ▸Architect and ship full-stack AI products end-to-end — scoping, system design, live deployment — for clients across UK, US, Canada, Dubai, and beyond.
- ▸Shipped a multi-workflow n8n property-management automation (Mocha Property, 10Ninety) — reducing manual processing from days to minutes.
- ▸Built self-hosted AI infrastructure on DigitalOcean with Docker, integrating OpenRouter, DeepSeek, and Anthropic APIs for live client deployments.
2024-01 — Present
Remote
Sole AI Engineer — Independent Contractor
SourceXAI
- ▸Sole engineer owning the complete AI product lifecycle — LLM integration, chatbot architecture, and automation design for international clients.
- ▸Built and documented 47 API endpoints for Phantm — a real-time AI avatar platform — using MuseTalk 1.5, CosyVoice 2, FastAPI, and RunPod.
- ▸Delivered CRM-integrated chatbots, voice agents, and intelligent workflow systems designed for reliability and scale.
2024-06 — Present
Remote
AI Trainer & Evaluator
Outlier · Mercor · Scale AI
- ▸AI evaluation and training across leading platforms: RLHF, preference ranking, code correctness, factuality, and instruction-following compliance.
- ▸Specialist domains: software engineering, AI & ML, full-stack development, prompt engineering, and STEM reasoning.
2023-01 — Present
Remote
Freelance AI Developer & AI Trainer
Independent — Global Clients
- ▸Completed 20+ AI and software projects for clients worldwide: mobile apps, SaaS builds, AI automation, and LLM integrations.
- ▸Built and published iOS and Android apps with full App Store deployment pipelines.
- ▸Developed n8n-powered lead-generation workflows integrating Apollo.io, Apify scraping, and AI-personalised email delivery at scale.
Live · grounded on the real résumé
Ask my AI twin
anything.
Built on Gemini, grounded on my projects and experience. Ask about Phantm, Mocha Property, my automation pipelines, timelines, availability — whatever you need to know.