SENTINEL
An on-device computer vision system, trained locally on Apple Silicon with MLX and YOLO26. Offline casualty triage — locally trained models, zero cloud dependency. Winner, webAI Build Challenge Enterprise Track.
ReadShort notes on artifacts that are already public — papers, datasets, public milestones, third-party verification letters, and the engineering decisions behind them.
An on-device computer vision system, trained locally on Apple Silicon with MLX and YOLO26. Offline casualty triage — locally trained models, zero cloud dependency. Winner, webAI Build Challenge Enterprise Track.
ReadThird-party verification letter from webAI confirming Igor Eduardo Borges as winner of the YOLO26 MLX Build Challenge 2026 Enterprise Track for SENTINEL.
ReadA research platform for peptide protocol tracking and real-world-evidence generation, built in 72 hours around interoperable research data. Winner, Antler ATX × Codex Hackathon.
ReadThird-party verification letter from AITX confirming the DeepInvent Best Science Hack result at the AITX Codex Hackathon for the submission Eterna.
ReadThe first benchmark for detecting Brazilian PII in Portuguese clinical text — seven identifier types with mod-11 checksum validation. The public paper, repository, and dataset — without restating unpublished claims.
ReadBM25 and dense retrieval as complementary methods, fused with Reciprocal Rank Fusion. The public paper, dataset, and engineering lab — without restating unpublished claims.
Read