determinism

The deterministic context engineering platform for open source AI. Connect open models and ontologies with context graph harnesses to build explainable, reliable agents.

ECS for Unity with full game state automatic rollbacks

Providing reproducibility in deep learning frameworks

Intuitive concurrent programming in any language

Smart Promise Tracker AI 2026 – Commit Memory Engine

Convex component for durably executing workflows.

When the stakes are high, intelligence is only half the equation - reliability is the other ⚠️

A neuro-symbolic runtime architecture (DIR) for Responsibility-Oriented Agents (ROA). By decoupling probabilistic LLM reasoning (User Space) from deterministic execution (Kernel Space), it ensures safe, invariant-driven, and auditable AI autonomy.

A determinizing tracer using Ptrace

Xronos revolutionizes robotics software — enabling reproducible behavior, timing control, and rapid root cause analysis through built-in observability.

Deterministic Simulation using ROS 2

renew engine: an AI-first game engine in Rust, built to be operated. Bit-deterministic fixed-timestep simulation in Q47.16 fixed point, Vulkan rendering through ash, and every capability driven from the CLI with schema-versioned JSON. 29 modular crates, desktop and mobile. Apache-2.0. Early development.

Experimental pedagogy: I don't teach, I learn and document along the way. Anyone can join.

UE5 Deterministic math library (fixed point)

The open benchmark for measuring the reliability, reproducibility, and determinism of AI context systems. Transparent metrics, reproducible tests, vendor-neutral results. Context Trust Levels CTL 0-4.

Determinism: Bit-identical outputs under identical inputs, configuration, and execution environment.

My journey to understand the meaning of life, consciousness, and free will

Write one C++20 compute flow. Run it on CPU, Metal, or Vulkan. Get bit-identical results—with replay, fixed-point math, and no silent fallback.

Composable, offline, stdlib-only verifier for re-derivable, attested artifacts — recompute the answer, don't trust the claim. Apache-2.0.

Test suites + deterministic Python for your agent skills' decision logic. Adversarial persona reviewers write the tests; the generated code must keep passing them — zero LLM calls at inference. Try: uvx temper-skills audit <skill.md>