Roadmap & Changelog

0.1 Blockers

What stands between today and the 0.1 design-partner cut:

• 📚 Polish documentation
• 💸 Cost UAT
• 🪟 Dashboard UAT
• 🛒 Finalize hev-shop and publish the repo
• 🏗️ Narrow cluster topology defaults

Up Next

Planned for 0.2:

• 🧭 Pinecone VectorStore backend — a second store kind
• 📊 Performance benchmarks — published pass-through overhead
• 🌱 Namespace init — first-time embed population
• 🔥 Trending searches — reduce-shaped UDFs
• 🔑 License key validation
• 🧳 Portable key bundle — stand up search-only edges

Later

• ♻️ Soft delete with TTL + restore
• 🧮 Gateway-side query embedding — warm model endpoint
• 🕰️ Temporal queries — as_of selector
• 🐇 Exact kNN result cache
• 🧪 A/B variant indexes
• 🦚 Per-query observability with LLM-judged Tail Quality
• 📣 Write amplification baselines
• 📮 layer push — Python UDF dev experience

0.1 Release (UAT)

API hardening

• 🧩 Finalize CRDs
• 🚆 Wire-compatible pass-through reads and writes
• 🏷️ Naming things

Lifecycle and operability

• 🎚️ Autoscaling compute for pipelines and UDFs
• 🗄️ Document cache endpoint for multi-stage pipelines
• 📸 Index snapshot history
• 🧨 Coordinated delete
• ⛵ Helm and Terraform install scripts
• 🔐 Scoped API keys — minted ApiKey resources
• 🏭 Warehouse CRD — declared Snowflake sources

Surfaces

• 🪟 Dashboard MVP — CRD management and observability
• 📚 Documentation site
• 🧰 Official Python, Go, and TypeScript clients

Search

• 🎯 Stable reads during heavy writes
• 🚦 Ready signal — namespace reports when every row is indexed
• 📜 Precomputed facet listings in snapshots
• 🪙 Precomputed facet counts in snapshots
• 🪃 Scans — row selection by filter, fts, or ann
• 🆔 Search by id via document-cached vector
• 🪢 Hybrid text fusion — fuzzy + BM25 fused by RRF
• 🧭 Query routing — Auto picks the search mode
• 📰 Search history saved to S3
• 🗂️ Enhanced namespace metadata