Supercharge Your Model Training

AIStore: scalable storage for AI applications

New file format for storage of large columnar datasets.

MONeT framework for reducing memory consumption of DNN training

GenAssist combines orchestration, runtime, analytics, and learning — in one open platform.

An MLOps workflow for training, inference, experiment tracking, model registry, and deployment.

Collection of OSS models that are containerized into a serving container

tracebloc notebook to launch and manage experiments in collaboration

Beamline is a tool for fast data generation for your AI/LLM/ML model training, simulation, and testing use-cases. It generates reproducible pseudo-random data using a stochastic approach and probability distributions, meaning you can create realistic datasets that follow specific mathematical patterns.

Integrating Aporia ML model monitoring into a Bodywork serving pipeline.

⌨️ Solutions to Academy Yandex "Тренировки по Machine Learning"

High-performance synthetic enterprise data generator. Produces 100+ interconnected financial tables — GL journal entries, document flows, subledgers, banking/KYC/AML, process mining (OCEL 2.0), graph exports (PyTorch Geometric, Neo4j), and 20+ process chains — with Benford's Law compliance, ACFE-aligned fraud labels, and formal privacy guarantees.

Topology-aware Kubernetes scheduler for multi-tenant, heterogeneous clusters

Self-Hosted MLFlow Docker Image with MySQL and S3 support

Smart Script to Mass Convert PDF .pdf to Markdown .md

Create a memorized array of unlimited numbers from a small seed. The output can be tokenized and used in code to derive values. Useful for synthetic data, personalization, world building and more.

Propensity model training with XGBoost

learning python day 4

Train a simple text classifier and predict labels - supports ONNX output for performance, language-neutral

MLflow adapter for CrateDB.