Portfolio
AI systems built for production
Six case studies across LLM applications, computer vision, document automation, and multimodal pipelines. Each includes architecture, metrics, and production considerations.
LLM Application
RAG chatbot for company knowledge bases
Upload internal documents and ask questions in plain English. Answers stream back in seconds with expandable source citations pinned to the exact page.
View case study →
Document Automation
Invoice OCR and document automation
From PDF to validated JSON in under 3 seconds: vendor, line items, totals, and due dates extracted and ready for your ERP or accounting system.
View case study →
Computer Vision
Retail shelf intelligence and inventory monitoring
Upload a shelf photo and get product counts, empty zone detection, and stock-level alerts — no manual audit needed.
View case study →
Infrastructure Intelligence
Road and infrastructure damage detection
Turn dashcam footage and drone imagery into a prioritised repair list, automatically detecting cracks, potholes, and surface damage in real time.
View case study →
Multimodal Pipeline
Video and audio intelligence system
Transcribe meetings and calls, auto-summarize, extract action items, and make any recorded content searchable in minutes.
LLM Engineering
LLM evaluation, fine-tuning, and structured output automation
Compare baseline prompting vs LoRA fine-tuning for consistent JSON classification — with full before/after evaluation metrics.