Pinned Loading

1. ContextOS ContextOS Public

   **Context Intelligence Operating System for LLMs.** ContextOS reduces the tokens you send to an LLM — cutting cost and latency — while keeping answer quality equal to or better than sending the ful…

   Python

2. Dev01 Dev01 Public

   A production-grade, multi-tenant AI console with sandboxed code execution, a universal connector system, streaming chat, and per-capability quality controls — all backed by Gemini and E2B.

   JavaScript

3. Aqueduct Aqueduct Public

   A distributed, durable task queue and scheduling engine, built from scratch in Go on top of PostgreSQL. At-least-once delivery, lease-based concurrency-safe dispatch, fault-tolerant workers, cron s…

   Go

4. HiveMind HiveMind Public

   Replace vector-DB RAG chunking with a self-organizing, file-based brain — an LLM parses incoming documents once at ingestion time and segregates information into topic-coherent .md files, which aut…

   Go

5. Astronoma Astronoma Public

   A single-node multi-tenant platform: S3-style resumable object storage, GCP-style IAM with user-composable custom roles, token-bucket rate limiting, write-behind caching, and streaming local AI inf…

   Python 1

6. ai-infrastructure-suite ai-infrastructure-suite Public

   System overview: a shared library (aikit) and three production-shaped AI-backend services built on it