A software ecosystem focused on low-overhead systems programming, high-throughput data pipelines, and raw machine learning preprocessing. Development centers on specialized, zero-dependency tools in C, C++ and Java where performance is entirely explicit.
Grease & Brine
Pinned Loading
Repositories
Showing 10 of 11 repositories
- dqn-ppo-crafter-paper Public
Evaluating Value-Based and Policy-Gradient Deep Reinforcement Learning Under Increasing Action-Space Complexity: A Crafter Navigation Study
- tally-sticks Public
The central repository for tracking and routing engineering epics across Grease & Brine.
- ne-pp Public
NE-PP (Neural Engine - Preprocess Pipeline), a C++ machine learning library implementing a modular preprocessing pipeline and neural engine, with native Python and Java bindings. Designed for performance-critical workloads with zero external ML dependencies.
- gnb-bucket Public
A consolidated monorepo workspace organizing focused topic buckets across software engineering and hardware, covering Arduino projects, computer science theory, LeetCode practices, and machine learning workflows.
- byte-modular-architecture Public
- gnb-management Public
Central repository for organization logistics, project tracking, and engineering coordination across all Grease & Brine initiatives.
- entropy-based-randomizer Public
Minimal C++ library for generating random numbers directly from system entropy (/dev/urandom on Linux/macOS, BCryptGenRandom on Windows). Header-only and cross-platform.
- .github Public
- mald Public
[Under Development] A real-time NLP classification system designed to identify and classify harmful, unsafe, or policy-violating text within conversational environments.
Top languages
Loading…
Most used topics
Loading…