YaFSDP: Yet another Fully Sharded Data Parallel

Main repository for Datadog Agent

tiktoken is a fast BPE tokeniser for use with OpenAI's models.

llama3 implementation one matrix multiplication at a time

Lecture notes for a short course on proving/programming in Coq via SSReflect.

Code for the book "The Elements of Differentiable Programming".

Distribute and run LLMs with a single file.

ClickHouse® is a real-time analytics DBMS

Shotizam analyzes the size of Go binaries

Neon: Serverless Postgres. We separated storage and compute to offer autoscaling, code-like database branching, and scale to zero.

The Orchestration Engine To Deliver Self-Service Infrastructure ⚡️

felixge's notes on the various go profiling methods that are available.

Get Things Done with Emacs

A high-performance observability data pipeline.

Store SSH keys in the Secure Enclave

A very minimal but elegant emacs (I think)

A low-latency, cloud-native KVS

A collection of TLA⁺ specifications of varying complexities

Interfacing R and Datadog

TLC is a model checker for specifications written in TLA+. The TLA+Toolbox is an IDE for TLA+.

Mirror of Apache Kafka

A hands-on-lab/tutorial for learning JDK Mission Control 7+.

Automated upstream mirror for libbpf stand-alone build.

A data-driven compute platform

a fast and light-weighted key-value store library

Distributed Stream Processing

NoSQL data store using the SEASTAR framework, compatible with Redis

Experimental, scalable, high performance HTTP server

extract internal monitoring data from application logs for collection in a timeseries database

Dogscaler scales up AWS autoscale groups based on the results of a datadog query.