ML Reliability · Agentic AI · LLM Systems · Graph Analytics · C++ / Python
Open-source projects built around measurable, reproducible evidence ↓
⭐ Featured — DeciShift
DeciShift is an open-source ML behavioral regression testing framework that compares two versions of a decision system, finds record-level action changes, attributes those changes to versioned components, preserves verifiable evidence, and can block a release when declared Decision Contracts are violated.
Public evaluation example: model accuracy improved from 90.26% → 91.79%, yet 30 of 719 final actions changed; a governed cohort exceeded its declared action-shift limit, so the release contract returned BLOCK.
ML Testing · MLOps · Model Governance · Decision Systems · Python
⭐ Explore / Star · PyPI · Live Demo · Dataset
pip install decishift
decishift demo --rows 1000 --no-save02 — 🧠 AgentWeave
Route before you reason.
Pre-inference routing and secure execution for tool-rich LLM and multi-agent systems. AgentWeave reduces the action space shown to a model before inference while keeping authorization, provenance and recovery explicit.
Project benchmark: 70.18% fewer tools exposed · 61.70% fewer input tokens · 50.95% lower mean local-model latency
MCP · A2A · LangGraph · AutoGen · Python
⭐ Explore / Star · PyPI · Docs · Paper
03 — ⚙️ MemVanta
Run quantized LLMs with less resident memory.
A memory-first C++20 runtime for quantized GGUF models on CPU, built around mmap-backed model access, paged KV cache, compact kernels and reproducible benchmarking.
7B benchmark: 3.80 GiB peak RSS · 47.54% lower peak RSS than the pinned comparison runtime
C++20 · GGUF · CPU Inference · Systems
⭐ Explore / Star · PyPI · Docs · Benchmark
04 — 🕸️ VeloGraphX
High-performance analytics for continuously evolving graphs.
A C++20 + Python engine for dynamic graph analytics with adaptive repair vs recomputation across BFS/SSSP, connected components, triangle counting, k-core and PageRank.
Retained exactness stress result: 2,000,000 updates · 0 BFS mismatches · 0 triangle mismatches
C++20 · Python · Dynamic Graphs · Graph Analytics · PyPI
⭐ Explore / Star · Docs · PyPI
I try to make projects useful beyond a demo: benchmarks, reproducible evidence, tests, releases, documentation and explicit claim boundaries alongside the code.
If one of these projects solves a problem you care about, star that project repository so others can discover it too. Issues, benchmark reproductions, integrations and technical feedback are also welcome.
Google Scholar · IEEE Xplore · ORCID · DBLP · OpenReview · ACM DL
Book: Machine Learning for Finance · Course: Data Analysis for Business and Finance · Technical writing: Towards Data Science, HackerNoon and KDnuggets
Research highlights · Research impact · Open-source contributions · Industry recognition
Follow @sauravsingla for releases, benchmarks and reproducible open-source experiments.