Model Training & Fine-tuning
Adapting models to the domain: LoRA/PEFT fine-tuning of vision-language models, quantization for serving, and classical ML — forecasting, recommendation and NLP.
Tool-Selection LLM — LoRA Fine-tune on xLAM
Teaching a small LLM to pick the right tool/API for an agent — and say how confident it is. LoRA-tuned Qwen3.5-2B with Unsloth on xLAM function-calling data: 50.6% → 100% accuracy, ECE 0.220 → 0.002.
3B Tokens/Day Vision-LLM Inference on One GPU
Scaled self-hosted vision-LLM screenshot analysis to 3 billion input and 50 million output tokens per day on a single 96GB GPU — no third-party API spend.
Vision-LLM LoRA Fine-tuning for Threat Triage
Fine-tuned a vision-language model with LoRA/PEFT to predict threat category directly from evidence screenshots, automating analyst triage.
Predictive Forecasting for a Banking Client
Forecasting models and automated reconciliation for a leading Australian bank — better accuracy than baseline and 40% less manual effort.