Clinton AI

Case Studies

Real projects, measurable results

Every project we take on has a clear problem, a thoughtful approach, and measurable outcomes. Here is a selection of our work across industries.

Building a Bloomberg-Style AI Sports Prediction Platform

Medallion Sports · Sports Technology

Sole AI Engineer & ArchitectDecember 2025 — present

Sole engineer on a full-stack AI prediction platform for NBA and PGA golf — combining real-time data feeds, machine learning models, prediction market analysis, and a conversational AI layer.

The Challenge

The client wanted a Bloomberg-style sports intelligence platform: real-time data, predictive models, market edge calculations, and actionable insights for subscribers. Nothing existed — it needed to be built from scratch, fast, and to commercial quality.

Our Approach

I took full technical ownership as the sole engineer. Designed the data architecture, built the ML pipeline, created the frontend dashboards, and deployed the entire system. The key architectural decision was separating the prediction engine (Python/FastAPI) from the presentation layer (React/TypeScript) to allow independent scaling.

What We Built

Built end-to-end: FastAPI backend with SQLAlchemy ORM, React frontend with real-time dashboards, Elo-based backtesting engine, market edge calculations using Kelly criterion, Polymarket integration for prediction market data, and Claude AI integration for natural language queries.

Results

  • 100+ daily predictions generated automatically
  • NBA MVP prediction model and PGA golf expansion launched
  • Live commercial product with paying subscribers
  • Full-stack platform designed, built, and deployed in under 3 months
  • Claude AI chat integration for conversational sports intelligence
PythonFastAPIReactTypeScriptPostgreSQLClaude APIStripeSupabase

Scaling Culturally Accurate AI Image Generation for Education

African Language House · Education Technology

AI Engineer & Pipeline ArchitectOctober 2025 — present

Designed and built an AI image generation pipeline that produces thousands of culturally appropriate educational illustrations for an African language learning platform — at a fraction of the cost of manual illustration.

The Challenge

The client needed 2,400+ educational images across 14+ categories — animals, food, household items, cultural scenes, body parts — all culturally appropriate for African language education. Manual illustration was prohibitively slow and expensive.

Our Approach

The core challenge was not image generation — it was cultural accuracy at scale. I designed a 5-tier prompt engineering system where each category has specific rules. Smart reference selection ensures style consistency across thousands of images.

What We Built

Built on Google Apps Script with Gemini 2.0 Flash: 14+ automatic category detection, 5-tier prompt engineering with category-specific cultural rules, queue-based batch processing with fault tolerance, smart reference selection, and automated Google Drive organisation.

Results

  • 2,422 total images planned, 646+ completed and delivered
  • 14+ automatic content category detection
  • 5-tier prompt engineering system for cultural accuracy
  • 80% reduction in image production time vs manual illustration
  • Fault-tolerant batch processing — resume where you left off
Google Apps ScriptGemini 2.0 FlashGoogle SheetsGoogle DrivePrompt Engineering

Zero Critical Issues: AI SaaS QA & Product Strategy

Partaake · Event Technology

QA Engineer & AI Product StrategistApril — October 2025

Led quality assurance and contributed AI product strategy for a golf event management SaaS platform. Caught 8 critical bugs pre-launch and helped shape the AI pricing model. Result: zero critical issues in production.

The Challenge

Partaake was preparing to launch its AI-powered golf event platform. The AI email generation system had silent metadata injection failures where personalisation data was being dropped without warning.

Our Approach

Rather than surface-level testing, I focused on understanding the AI system's failure modes. I traced the email generation pipeline end-to-end, identified the exact data conditions causing silent failures, and documented structured reproduction steps.

What We Built

Conducted systematic QA across the entire platform with deep focus on the AI email system. Logged 30+ structured issues in Linear. Caught 8 critical bugs pre-launch. Contributed pricing strategy: proposed tiered AI usage caps to protect margins at scale.

Results

  • 30+ structured issues documented in Linear with reproduction steps
  • 8 critical bugs caught and resolved before launch
  • Zero critical post-launch issues — clean launch April 2025
  • AI pricing strategy (tiered usage caps) adopted into product
  • AI email system launched with proper fallback handling

Clinton's testing went far beyond what I expected. He didn't just find bugs — he understood the AI system deeply enough to identify systemic issues and propose solutions.

Kat Slump

Founder & CEO, Partaake

LinearAI Email SystemsSaaS TestingProduct StrategyPricing

Enterprise AI Evaluation: From Inconsistent to Reliable

Velocity AI · Enterprise AI / Research Intelligence

AI Engineering ArchitectMay 2025 — present

Built the multi-model LLM pipeline, evaluation framework, and guardrail system that became the backbone of an enterprise AI research company.

The Challenge

Velocity AI was generating research insights for enterprise clients using LLMs, but output quality was inconsistent. There was no evaluation methodology, no quality benchmarks, and no way to catch hallucinations.

Our Approach

The fundamental insight was that enterprise AI needs the same rigour as enterprise software: testing, benchmarking, regression detection, and quality gates.

What We Built

Designed a multi-model orchestration pipeline (ChatGPT, Perplexity, Grok) with cross-model verification. Built the evaluation methodology from scratch: golden datasets, regression tests, red-teaming protocols, and structured quality scoring.

Results

  • 40% reduction in research delivery time
  • 30% improvement in factual reliability of outputs
  • Enterprise-grade evaluation framework still in use
  • Ongoing engagement since May 2025 — now core infrastructure
  • Zero hallucination incidents on evaluated outputs
PythonOpenAIPerplexityGrokEvaluation FrameworksGolden Datasets

First AI-Powered Maritime Procurement System

PortContact · Maritime & Shipping

Sole AI Engineer & ArchitectJanuary 2026 — present

Built the first tool in the maritime industry to generate RFQ content using AI. A 3-stage pipeline that replaced hours of manual procurement work with accurate, validated output in minutes.

The Challenge

Maritime procurement officers spent hours manually creating RFQ documents by cross-referencing catalogues, specifications, and supplier data. No tool existed to automate maritime procurement content generation.

Our Approach

The critical design decision was JSON-first architecture. In maritime procurement, a wrong part number or specification can cost thousands. I designed the system so the AI never generates numbers or specs from memory.

What We Built

Designed a 3-stage AI pipeline: Analyst (parses requirements), Generator (creates RFQ content from 500+ real catalogue items), Validator (checks every output). JSON-first architecture ensures zero hallucinations on critical procurement data. 22 API endpoints.

Results

  • First tool to generate maritime RFQ content with AI
  • 500+ catalogue items indexed and queryable in real-time
  • 22 API endpoints in production
  • Zero-manual-QA: every output structurally validated
  • Hours of manual procurement work reduced to minutes
Next.js 15TypeScriptOpenAISupabasePDF ProcessingREST APIs

Have a project that needs AI expertise?

We bring the same rigour and production-grade engineering to every project we take on.

Start a Conversation