The work behind the flagship

Privacy-preserving synthetic training data

Synthetic data generation for model training in regulated domains where real data cannot leave the building. Customer data schemas stay on-premises; only statistically equivalent synthetic outputs are used for training. Feeds the same boundary-first thesis as Argus.

Structured multi-model reasoning

Two model instances argue opposing positions on a given thesis with rolling summarisation for extended exchanges. The structured-disagreement machinery behind Argus's assessment council began here.

Adaptive dictation that learns how you speak

Developer dictation powered by local Whisper transcription and a layered correction pipeline that improves with every fix. Accent-tolerant, jargon-aware, entirely on-device: no audio leaves your machine. Named from the Icelandic 'tala', to speak.

Multilingual dictation for Arabic code-switching

Real-time speech-to-text for Arabic speakers who naturally switch between Arabic, English, and French mid-sentence. Streaming API with dialect-aware recognition, bidirectional text rendering, and post-processing on our own inference stack.

Language preservation research

Exploratory work on how AI, corpus methods, and digital tools can support endangered and historical languages. Research notes and framework development, published as the project takes shape.

Learn Spanish in your own voice

A lab exploration in voice cloning and neural translation: speak English, hear yourself in Spanish. Voice clones are discarded after use. A demonstration of the speech stack rather than a product commitment.

Infrastructure that diagnoses, heals, and remembers

The operating layer underneath everything we run: continuous low-cost triage across every signal, escalation of novel failures to a deep reasoning tier, and remediation committed as permanent immunity. The perception, orchestration, and memory layers (developed as pNexus and pureMind) ship as parts of one system.

Sentinel's orchestration layer

The coordination layer of the Sentinel system: multi-tier inference routing, task scheduling, memory consolidation, and tool integration across the fleet. Documented separately for its engineering history.

Sentinel's memory layer

Git-versioned knowledge base with hybrid semantic and full-text retrieval, automated consolidation, and programmatic scheduling. The memory that lets Sentinel turn incidents into immunity.

Edge perception exploration

A wall-mounted sensor unit exploring lightweight local perception: camera, microphone array, on-device transcription, with heavy reasoning handed off to the fleet.

GPU-accelerated 3D reconstruction

3D Gaussian Splatting and Neural Radiance Fields on the fleet's GPUs. Reconstruction experiments from photo and video input, kept as a lab note.