_peregrine_
- Karma
- 0
- Created
- ()
- Submissions
- 0
- A graph-theoretic approach to building reliable LLM judges for retrieval (georgianailab.substack.com)
- How to build a distributed queue in a single JSON file on object storage (turbopuffer.com)
- ANN v3: 200ms p99 query latency over 100B vectors (turbopuffer.com)
- Designing inverted indexes in a KV-store on object storage (turbopuffer.com)
- Vectorized MAXSCORE over WAND, especially for long LLM-generated queries (turbopuffer.com)
- Flink's 95% Problem (tinybird.co)
- How to ingest 1B rows/s in ClickHouse (tinybird.co)
- How we built our own Claude Code (for data) (tinybird.co)
- Show HN: Click the Circle (tb-peregrine.github.io)
- Tinybird made a ClickHouse CLI agent (tinybird.co)
- Why LLMs Struggle with Analytics (tinybird.co)
- Claude Opus 4 is the best model at SQL generation (llm-benchmark.tinybird.live)
- We graded 19 LLMs on SQL. You graded us (tinybird.co)
- Run Tinybird on your own infra (for free) (tinybird.co)
- Tinybird Forward Hackathon (github.com)
- Forward: Evolving Tinybird for the AI-native developer (tinybird.co)
- Real-time logs analytics architectures that scale (tinybird.co)
- Analyzing the Dub.co Analytics Playbook (tinybird.co)
- AI Engineer Starter Pack – Volume 2 (aiengineerpack.com)
- How to analyze MCP server usage (tinybird.co)