This repository is an index of selected research systems and open-source implementations developed under my technical leadership and supervision. These works form part of the foundation for my current work as founder and Chief Scientist at EyeTrustAI (AI decision control under uncertainty).
Primary reference (context + narrative): 👉 https://www.alejandrosanchezguinea.com
Note on attribution: implementations are hosted in the repositories of the primary developers. I do not mirror or fork student code here; links below point to the original sources.
- Repo: https://gitlab.com/achref.d/nesymof
- Focus: neuro-symbolic forecasting with explicit structure and uncertainty-aware reasoning.
- Repo: https://gitlab.com/achref.d/clear-command
- Focus: LLM-based emergency-response coordination with decision/control structure.
- Repo: https://gitlab.com/achref.d/krps
- Focus: structured reasoning and planning components for uncertainty-aware systems.
- Repo: https://github.com/thkreutz/umosmots
- Focus: 4D LiDAR motion segmentation in stationary settings via multivariate occupancy time series.
- Repo: https://github.com/thkreutz/crowdorchestrationsim
- Focus: simulation framework for spatio-temporal spawn dynamics and crowd orchestration.
- Repo: https://github.com/thkreutz/despite
- Focus: contrastive multimodal embeddings for human activity understanding.
Across these projects, my contributions typically focused on:
- defining the system and decision architecture
- specifying representations (uncertainty, multimodal structure, neuro-symbolic components)
- shaping evaluation and validation strategy for safety-critical contexts
- supervision, research direction, and publication strategy