DEIC research prototype for hidden-state inference, source reliability, diagnostic querying, and adaptive recovery.
-
Updated
Jun 14, 2026 - Python
DEIC research prototype for hidden-state inference, source reliability, diagnostic querying, and adaptive recovery.
Next-word prediction web app: an LSTM trained on ~3,000 quotes (Keras/TensorFlow), served with FastAPI on Render, with a vanilla HTML/CSS/JS frontend that handles cold-start health checks.
Not-peer-reviewed SPPT/ASTRA planetary-systems framework with accessible versioned editions, synthetic demonstrations, and separate Earth-systems working papers.
To associate your repository with the hidden-state topic, visit your repo's landing page and select "manage topics."