This repository implements generative recommendation models, primarily HSTU (Hierarchical Sequential Transduction Units) and DLRM-v3, for large-scale sequential recommendation. It is a hybrid codebase combining research experimentation (ICML'24 paper on generative recommenders) with production-grade inference benchmarking (MLPerf loadgen integration). The code provides custom high-performance CUDA and Triton kernels for flash attention, jagged tensor operations, and layer normalization, targeting modern NVIDIA GPUs (Hopper/Blackwell). It includes training pipelines with gin configuration for datasets like MovieLens, Amazon Books, KuaiRand, and synthetic streaming data up to 100B items. The repository also contains a full MLPerf inference benchmark suite with vendored loadgen library and end-to-end TorchScript tracing for deployment. While the core focus is research on scaling laws and architecture innovations for sequential recommenders, the ops and modules are engineered to production standards with multiple kernel backends and rigorous tests.
This repository is the official Algolia API client library for Go (v4). It is a generated SDK that provides typed Go structs, serialization, and HTTP transport for interacting with all Algolia services including Search, Analytics, Recommendations, Ingestion, A/B Testing, Personalization, Monitoring, Insights, Agent Studio, and Query Suggestions. The code is overwhelmingly auto-generated by OpenAPI Generator from Algolia's shared API specs, with a hand-written core infrastructure layer for HTTP transport (rate-limiting, retry with host failover, stateful host management) and utilities (nullable types, deep merge, SSE streaming). The library is well-structured and follows consistent patterns across all service packages, but the heavy reliance on code generation leads to significant duplication of identical models across packages and a large number of boilerplate files. The hand-written transport layer is robust and well-tested, demonstrating good engineering practices. This SDK is suitable for any Go application needing to integrate with Algolia's search and data platforms.