Neuromorphic Inference Lab
Formal framework for persistent LLM agents with welfare anchored to externally monitored world-states. Interventional hope term via do-calculus. Governance-transparent utility decomposition.
A persistent agent whose primary welfare is world-directed. Monitors PM2.5 air quality in real time, deliberates on action selection, and exposes its full thought stream as it happens.
Medium-voltage fault risk inference with CI-minded training flow, artefact versioning, and live API serving.
Forecasting as a system: feature pipelines, backtesting harness, and deployment-ready outputs for planning workflows.
Retrieval tracing with source citations, guardrails enforcement, and automated evaluation harness for LLMOps workflows.
Complete MLOps pipeline: data ingestion → training → inference API → monitoring → CI quality gates.
Interactive tabular risk scoring with client-side baseline model and optional API integration for deployed endpoints.
Image analysis tool with local property inspection and optional API integration for deployed vision models.
AI-powered transaction categorisation using NLP to extract merchant names, categories, and anomaly signals from raw Open Banking data.
Zero-shot NER microservice for banking documents. Detects PII and financial identifiers using a DeBERTa encoder quantised to INT8 for CPU serving.