Category

MemPalace

Stores conversation history verbatim and retrieves it via local semantic search with a structured index (wings/rooms/drawers). Pluggable vector backends and a local-first default mean high recall (benchmarked) without cloud or API keys—useful for agent memory and private RAG.

CubeSandbox

Provides hardware-isolated, sub-60ms, ultra-low-overhead sandboxes to run untrusted LLM/agent code. Offers event-level snapshots, kernel-level egress control, credential vaulting, and drop-in E2B SDK compatibility for high-density AI agent deployment.

Browser Harness

Connects an LLM to a real browser over an editable CDP websocket so the agent can drive clicks, navigation, and generate missing helper code during tasks. The harness self-heals by writing reusable helpers, supports local or cloud browsers, and can optionally record sessions for debugging.

ADR

Monitors and detects risky behavior in enterprise AI agents via high-fidelity telemetry, security benchmarking, and a two-tier detector. Comprises ADR Sensor, ADR-Bench, and ADR Detector; deployed in production at Uber and validated on public benchmarks.

FlashKDA

Provides high-performance CUDA/CUTLASS kernels implementing Kimi Delta Attention (KDA), accelerating KDA prefill on SM90+ (Hopper) GPUs. Integrates as a drop-in backend for flash-linear-attention, supports native variable-length batching, and targets K=V=128; requires CUDA 12.9+/PyTorch 2.4+.

Future AGI

Provides an end-to-end platform to evaluate, observe, protect, and optimize LLM and AI agent deployments. Integrates OpenTelemetry tracing, 50+ evaluation metrics, agent simulations, an OpenAI‑compatible gateway, and guardrails; self‑hostable under Apache 2.0.

AiSOC

Ingests and normalizes security telemetry, runs multi-model AI agents to produce replayable investigations and automated triage/response; key features include a step-by-step Investigation Ledger, CI-gated eval harness, and self-hostable deployments.

Knowledge Catalog

Provides tools and samples to build context management, enrichment, and retrieval solutions on Google Cloud Knowledge Catalog — an AI-oriented data catalog that builds a dynamic knowledge graph for structured and unstructured data, suitable for RAG and agent workflows.

OpenTelemetry GenAI Semantic Conventions

Defines OpenTelemetry semantic conventions for generative AI telemetry — spans, metrics, and events for GenAI clients, the Model Context Protocol (MCP), and provider-specific integrations. Includes YAML models, human-readable docs, and reference implementations to standardize observability across GenAI deployments.

DwarfStar 4

Native local inference engine for DeepSeek V4 Flash (also supports GLM 5.2 and PRO on high‑memory machines). Focused features include model-specific loading, SSD expert streaming, asymmetric routed-expert 2-bit quant support, multi-GPU/tensor/pipeline parallelism, and an OpenAI-compatible server plus a native coding agent.

TokenSpeed

High-throughput LLM inference engine for agentic workloads, combining a local‑SPMD static compiler for parallelism, a C++ scheduler with a Python execution plane and type‑safe KV‑cache reuse, pluggable high-performance kernels (including an MLA implementation), and a low‑overhead AsyncLLM entrypoint for production GPU inference.

Agent Substrate

Runs a Kubernetes-native runtime that multiplexes many stateful agent-like actors onto a small pool of sandboxed worker pods via full-state snapshots and pre-warmed workers, enabling sub-second suspend/resume and 30x+ oversubscription.