evidence models for physical systems
projects · Klosr · Orca Strike · repo · contact
aimez builds evidence models and tools for decisions about complex physical systems. The work turns sparse observations, physical constraints and audit records into representations that can be inspected before they guide a route, forecast or research artifact.
The company works on the evidence gap between complex physical environments and the records available when people make decisions. Each project starts with a concrete representation problem, then publishes the artifact, paper or public module that exposes the boundary of the evidence.
Klosr is a city-scale evidence model for pedestrian conditions on Manhattan, where sparse observations fuse onto a walk graph and validated evidence layers combine into a route objective.
Comfort-aware routing is one application on that representation, while five cost encodings read the same evidence layers to compare shortest, crowd-avoiding, shade-seeking, comfort-aware and adaptive routes. Each narrated value carries a live, partially live or deferred provenance label.
A pedestrian camera network that samples every 30 minutes builds a detailed record of street conditions, but that record is always of conditions 30 minutes ago. When the 08:00 cycle closes, it contains what the cameras observed from 07:30 to 08:00, while a decision that depends on conditions at 08:30 has nothing measured to consult. Field-JEPA is an experiment on forecasting the next crowding field from that measured cycle without waiting for the next cycle to close.
The live application hosts remain in limited access while the public release prepares, so the public surface below is the Field-JEPA holdout timelapse. Research notes stay secondary while the release prepares.
Field-JEPA forecast · Neighborhood access research (archive)
orcast is an evidence-gated encounter forecasting research platform for Southern Resident killer whales in the Salish Sea. A kernel intensity model with validation gates and provenance keeps displayed confidence tied to held-out skill, while the research workbench covers biologging-driven 3D reenactment and hydrophone views with a public launch planned for this summer.
Application hosts remain in limited access until that release, so the public module embedded below is the Orca Strike interaction surface.
Research archive orcast whitepaper 1 (PDF) · orcast whitepaper 2 (PDF)
wavves supports durable multi-agent routing so long-running sessions continue from recorded files instead of chat memory. Lanes advance only after a disk gate capture records a pass or block verdict.
The Field-JEPA research panel for Klosr and the Orca Strike module for orcast remain available here while their full application hosts stay in limited access.
Beating persistence on Manhattan’s next half-hour of crowding, this holdout-day timelapse places continuous Field-JEPA forecasts beside observed crowding-sigma flowfields for the same cycle, with day of week and local time ticking between the panels and color encoding crowding sigma as a percentile rank of bias-corrected density inside a validated operating region.
On the full-island B cohort (339 cameras), Field-JEPA reaches 0.090 MAE at +30 minutes against a persistence bar of 0.105, a 14.2% reduction on a single held-out day (2026-06-29). The sigma field predicted here is a research result on Klosr’s crowding evidence layer, not a deployed routing feature.
Persistence, Field-JEPA and observed crowding sigma at +30 minutes appear as a continuous shoreline-clipped flowfield for a single holdout cycle, with persistence carrying the last observed field, Field-JEPA showing the forecast and the holdout panel showing what was observed.
Route overviews for two origin-destination pairs show the Village dog-run corridor and Midtown Grand Central to Carnegie Hall, with field-validated shade comparison scenes kept in the research pack as modeled shade and surface temperature on the street surface.
Fork-metric cards that carried calibration constants were removed from the public surface, while the aggregate Field-JEPA evaluation numbers above remain as research results.
Manhattan is the pilot city-scale evidence model for pedestrian conditions on a walk graph. Camera-measured crowding cycles sit beside modeled shade and other evidence layers. Field-JEPA adds a predicted crowding layer beside the current-cycle measured field.
orcast publishes gate-bounded encounter forecasting and claim-to-evidence binding for wildlife observation data, with Orca Strike as the current public interaction surface.
Developer tooling work studies moderator-led agent sessions, explicit handoffs and check records for long-running research work.
Research archive orcast whitepaper (PDF)
Occasional notes cover demos, papers and public artifacts from the aimez research program.
Join research updatesUse this channel for project inquiries, institutional contact, event invitations and research collaboration.
Start a conversationBroader diagnostic-infrastructure concepts in the research program are the subject of a filed U.S. provisional patent application, while the public materials on this page are a research demonstration under that pending filing.