Catalogue
- Cards
- 165
- Records
- 486
- Workloads
- 19
- Corpus version
- 74847648e25a
40 of 486 records match; the first 20 are shown.
| Kind | Record | Organisation | Year | Setup |
|---|---|---|---|---|
| Decision | Partition catalog clustering to per-shop tasks and filter standalone stores deterministically via Liquid themes before calling LLMs | Shopify | 2026 | |
| Decision | Replace naive linkage-tree clustering with FAISS HNSW plus sparse UPGMA average linkage to form coherent chunks up to 200 products | Shopify | 2026 | |
| Decision | Use a two-stage propose-and-critique LLM flow with dynamic per-chunk Structured Output schemas and normalized product IDs | Shopify | 2026 | |
| Decision | Pair GPT-4 confirmation and MetaReflection hybrid prompts with expanded offline evaluation to curb LLM false positives | GitHub | 2025 | |
| Decision | Share LLM capacity across push and full-history scans with dynamic rate-limit ranges and filepath exclusions | GitHub | 2025 | |
| Decision | Validate prompt and filter changes via mirror testing on public-preview repositories before GA | GitHub | 2025 | |
| Decision | Use few-shot GPT-4-chat (gpt-4-32k) instead of fine-tuning davinci-002 or using the completion API | Canva | 2023 | |
| Decision | Redact internal data pre-LLM, use soft 600-character prompt targets with retries, and track edits via Jira webhooks | Canva | 2023 | |
| Decision | Run nearest-neighbor vector search in memory inside stateless Kubernetes services rather than a stateful database | Spotify | 2023 | |
| Decision | Build Voyager as a new HNSW package to succeed Annoy instead of making breaking changes in hnswlib | Spotify | 2023 | |
| Decision | Replace the Wide & Deep (XGBoost + DNN) fraud ensemble with a ResNeXt-inspired multi-branch pure DNN | Stripe | 2023 | |
| Decision | Pair black-box DNN fraud scoring with merchant Risk Insights (backed by Elasticsearch) and internal feature-contribution tables | Stripe | 2023 | |
| Decision | Move zone-level parallel ML training from a shared long-running Celery queue to per-application Ray clusters on EKS | Instacart | 2023 | |
| Decision | Isolate Python environments per Ray cluster behind a unified Griffin launcher API for dev and Airflow runs | Instacart | 2023 | |
| Decision | Restrict Venice to asynchronous batch and stream writes with background version swaps, server-side partial updates, and hybrid replay | 2022 | Multi-region | |
| Decision | Push vector math onto Venice servers with Read Compute and embed Da Vinci as an eager local-state client library | 2022 | Multi-region | |
| Decision | Swap BDB-JE for pluggable RocksDB storage formats and operator-driven Zstandard shared-dictionary compression | 2022 | Multi-region | |
| Decision | Run active-active multi-region replication and rack-aware Apache Helix self-healing reusing the Full Push rebalance path | 2022 | Multi-region | |
| Decision | Enforce per-dataset storage and read quotas on multi-tenant clusters and group over-replicated partitions to bound fanout | 2022 | Multi-region | |
| Decision | Remove feed-mixer and interest-discovery dependencies by moving Post search fanout into the search federator | 2022 |
20 more records. Sign in to see all 40.