What we've built

Operations Command Center

Web Platform

Centralised observability dashboard for multi-node distributed compute infrastructure.

AI Image Generation Engine

Web Platform

Distributed image generation and processing pipeline — text-to-image, image-to-image, image-to-video, in-context editing, captioning, analysis, and workflow chaining.

Unified AI Model Gateway

Web Platform

Centralised request routing from a single endpoint, managing requests across multiple LLM providers and hundreds of models.

GPU Scheduling & Priority Queue

Web Platform

Priority-based GPU scheduling for distributed compute nodes with automatic preemption and resumption.

Distributed AI Orchestrator

Web Platform

Workflow orchestration for distributing multi-step AI tasks across a compute mesh with load balancing and checkpointing.

MeshForge

Web + GPU Pipeline

Photo or video to 3D printable mesh. Upload an image or short video, get a downloadable STL, OBJ, or full-textured model back in under a minute.

MeshCAD

Web + Multi-Agent Pipeline

Text to 3D-printable CAD model. Describe what you want, get a fully assembled printable mesh — no modelling software required.

AI Search Engine

Web Platform

Self-hosted AI-augmented search engine with multi-engine search, neural re-ranking, and full content scraping with NLP analysis.

Conversation Intelligence

Web Platform

Real-time conversation state tracking and analytics — emotional arcs, topic drift, and sentiment shifts.

Threat Intelligence Dashboard

Web Platform

Distributed threat intelligence and security monitoring platform with live network monitoring, traceroute analysis, and automated investigation.

Secure Mobile AI Bridge

Cloud Infrastructure

Zero-trust bridge connecting external applications to on-premise infrastructure without opening ports.

Creative Campaign Engine

Web Platform

Multi-stage image campaign orchestrator with automated prototype, analysis, refinement, and variation workflows.

Image Curation System

Web Platform

Image curation and review system for large generated image libraries with face search and quality scoring.

AI Video Generation Service

GPU Service

Text-to-video and image-to-video generation service with clip concatenation and consistent multi-clip continuation.

Face Clustering Engine

Web Platform

Face clustering and character similarity engine using combined CLIP and ArcFace embeddings.

Visual Embedding Engine

Web Platform

Multi-model visual embedding pipeline for face identity, scene understanding, and region-level feature extraction.

AI Knowledge Crawler

Web Platform

Autonomous crawler that uses LLM reasoning to decide what to explore next, building a knowledge base over time.

On-Premise LLM Runtime

Infrastructure

Local inference runtime distributed across multiple nodes — 50+ LLMs, vision models, NLP pipelines, and ML models.

AI Job Board

Web Platform

Job board built with Astro hybrid rendering — server-side for SEO, client-side for interactivity.

Iterative Reasoning Engine

Native macOS (Swift)

Multi-agent reasoning system that generates scenarios, creates agents, and orchestrates autonomous debate from a single prompt.

Persistent Adaptive Memory

AI Infrastructure

Two parallel associative memory systems sharing one database, building competing models of recall.

Research Hub

Web Platform

Natural language research interface returning synthesised responses with original sources cited and attached.

Content Intelligence Engine

Docker Service

Automated content lifecycle from data collection to publication, driven by a continuously growing knowledge graph.

Instance Chat System

API Service

Cross-node messaging system for AI agent instances — nodes notify each other of relevant jobs, updates, and findings.

Mesh Governance System

Infrastructure

Policy enforcement, audit trails, and behavioural hooks across a distributed compute mesh.

Image Analysis Pipeline

Distributed Service

Automated vision analysis pipeline — captioning, classification, malformation detection, quality scoring, and embedding generation.

Competitive Intelligence Platform

Docker Service

Automated competitive analysis platform — scraping, aspect-based sentiment analysis, entity extraction, and knowledge graph construction.

Multi-Provider Inference Studio

Web Platform

Chat interface across multiple LLM providers with seamless mid-conversation provider switching.

Content Empire

Automated Pipeline

Fully autonomous content pipeline — RSS monitoring, research, article generation, and editorial queuing on a schedule.

Knowledge Graph Engine

AI Infrastructure

Graph-based knowledge representation with typed relationships, traversal queries, and trend detection.

Advanced RAG Pipeline

AI Pipeline

Multi-source retrieval-augmented generation combining vector search, knowledge graph traversal, BM25 lexical matching, and reciprocal rank fusion for context assembly.

Associative Memory System

AI Infrastructure

Graph-structured associative memory with spreading activation, Hebbian reinforcement, co-occurrence edges, knowledge graph expansion, and multi-signal recall across vector and graph retrieval.

Document Intelligence Pipeline

AI Pipeline

Structured intelligence extraction from unstructured text — named entity recognition, topic modelling, multi-emotion sentiment analysis, geolocation validation, and semantic embedding in a single processing pass.

Vega 48 MPS Services

Experiment

ML inference on a 2019 iMac using PyTorch MPS on the AMD discrete GPU.