Central Casting
Checkpoint-based orchestration and audit for long, multi-repository agent work.
ExploreResearch focus
I build computational infrastructure for multi-scale analysis across behavioral and sensor-driven systems, with an emphasis on representational coherence under variation in scale, resolution and structure.
The work blends neuroethology, signal processing and simulation with a pragmatic goal: keep interpretability as internal descriptions shift.
Implementation landscape
Active themes I am implementing across projects. A working map of what I build and test, not a retrospective summary.
Systems and behavioral neuroscience
- Sensory-motor integration and behavioral outputs
- Navigation and spatial cognition in three-dimensional environments
- Attention, learning and memory in adaptive behavior
- Cross-species behavioral modeling and trajectory analysis
Theoretical neuroscience
- Representational geometry and predictive spatial maps
- Social perception
Machine learning and methods
- Adaptive windowing and feature extraction from complex signal data
- Memory-efficient recurrent architectures for time series classification
- Dynamic state-space modeling for behavioral transitions
- Multimodal representation of continuous signal patterns
Framework implementations
aimez.ai
Applied research program that takes public traffic-camera frames through computer-vision models into a measured stress field and into pedestrian routing that respects lived burden. It carries the work from raw field data to a public site with notes, figures and validation briefs.
Central Casting
Checkpoint-based orchestration and audit for long, multi-repository agent work. Agent homes act as role contracts, checkpoints are the unit of memory and an alignment check flags when the structured record and the human-readable views diverge.
Current work
- Aimez.ai (pax-nyc): Multi-scale signal processing for learned heuristic pathfinding in urban sensing, integrating multi-modal features from traffic cameras.
- magniphyq: Modular research visual workspace, a node graph that wires data, analysis and simulation into one runnable, inspectable pipeline, deployed as a live demo.
- Drosophila sensory circuit analysis: Pipelines and visualization tools for sensory integration across gustatory, olfactory, mechanosensory and visual modalities, with temporal pattern extraction from trajectory data.
- Marine behavioral states: Lightweight recurrent modeling for humpback whale behavioral states from dive telemetry, targeting multi-scale segmentation and event detection.
- INDYsim: Hierarchical simulation framework for behavior as the interface between perception and adaptation, using agent based and event driven modeling of transitions under constraints.
Projects
retrovibez
Larval reverse-crawl detection adopted by a mechanosensation group. Klein et al. 2015 method.
Exploremagatfairy
One-command bridge from legacy MATLAB analyzer data to HDF5 and Python, with validation and a CLI.
Exploremagniphyq
Modular research visual workspace: a node graph wiring data, analysis and simulation, with a live demo.
Explorewhale behavior analysis
Hierarchical classification of humpback foraging strategy from dive telemetry.
ExploreResearch program gallery
Public pages on aimez.ai/public. Each card opens a shareable research page (not private working notes). Hover for a short description.
Central Casting
An agent-run practice
A recent method for keeping multi-repository agent work coherent: checkpoints as memory, agent homes as role contracts and a deliberate local and cloud boundary.
magniphyqModular research visual workspace
A node-based workspace that wires data, analysis and simulation into one runnable graph: track extraction, segmentation, contour analysis and a synapsin phase diagram.
Live demoStress-aware routing
Interactive corridor routing: compare shortest vs stress-aware paths per leg or on a multi-stop walk.
Research overviewGraph-field research
Measured stress fields, directed edges, route cost, loop behavior and graph-level structure.
Mathematical frameCorridor topology primer
Paths, loops, and residuals as modeling language for the Manhattan graph work.
Spatial mediaInstallation sketch
Room-scale installation concept from camera-derived fields, light, route traces, and movement.
aimez
aimez is an applied research initiative investigating coherence in complex information environments. Many of the phenomena of interest behave as dynamical systems, where small changes in structure or scale can induce qualitative shifts in behavior. The current focus is the measured-city work: a camera-derived stress field and stress-aware pedestrian routing.
magniphyq
magniphyq is a modular research visual workspace where a whole pipeline lives as a node graph a researcher can run, inspect and clone. It wires track extraction, behavioral segmentation and contour analysis into one runnable graph, and it hosts a synapsin condensate phase-diagram campaign, so a measured model runs and compares against data in the same place. It is built and verified, deployed as a live demo.
Research papers
Collaborators and outreach
Researchers and labs whose work intersects with these themes. If you want to reach out, learn more or work together, leave your email and I will follow up.