Built to investigate. Engineered to be used.
Open software implementations that support empirical inquiry and put research methods to work. Organized by operational domain, with implementation maturity clearly documented.
Observe & Understand
Make the execution state and divergence of complex AI workflows inspectable.
AgentWatch
Runtime evidence, provenance, comparison, and replay for AI-native systems.
PulseStack
Observability and execution replay for distributed AI workflows and agents.
Build & Orchestrate
Connect empirical research questions to reproducible, high-throughput software pipelines.
RepoPilot
A repository agent that edits code, runs tests, and iterates on execution feedback.
VEX
Content-hashed incremental builds with task skipping and a shared remote cache.
FluxCore
Event-driven orchestration for reactive pipelines, services, and automation.
Explore & Apply
Carry experimental algorithms into working developer software and sensory runtimes.
MEOWCODE
A unified AI workspace that routes requests across configured providers.
MYCELIUM
An evolutionary runtime for generating, mutating, and evaluating executable software.
SentinelCV
Computer vision for fall detection and safety monitoring with CCTV and IP cameras.
LCI (Lifecycle Integration)
A public LCI repository was not found in the organization during our 11 October 2026 review. Its scope and implementation status remain unverified; therefore, no unsubstantiated claims or repository targets are published.
Good questions deserve independent reproduction.
Reproduce an experiment, challenge an assumption, or join the discussion with fellow researchers on Discord and GitHub.