# Hotdata > Give every agent an isolated, ephemeral environment to search, query, join, and analyze data without waiting on shared compute. Run SQL, vector, full-text, and geospatial workloads across your existing data platform. Every page below is the Markdown version of a page on https://www.hotdata.dev: the same URL with `.md` appended. https://www.hotdata.dev/llms-full.txt holds all of them in one file. ## Getting Started - [Quick Start](https://www.hotdata.dev/docs/quick-start.md) - [Core Concepts](https://www.hotdata.dev/docs/core-concepts-overview.md) - [Data Sources](https://www.hotdata.dev/docs/data-sources.md) ## Reference - [CLI Reference](https://www.hotdata.dev/docs/cli-reference.md) - [SQL Reference](https://www.hotdata.dev/docs/sql.md) - [Math functions](https://www.hotdata.dev/docs/sql-functions-math.md) - [String functions](https://www.hotdata.dev/docs/sql-functions-string.md) - [Date & time functions](https://www.hotdata.dev/docs/sql-functions-datetime.md) - [Conditional & utility functions](https://www.hotdata.dev/docs/sql-functions-conditional.md) - [Aggregate functions](https://www.hotdata.dev/docs/sql-functions-aggregate.md) - [Window functions](https://www.hotdata.dev/docs/sql-functions-window.md) - [Array, struct & map functions](https://www.hotdata.dev/docs/sql-functions-array.md) - [JSON functions](https://www.hotdata.dev/docs/sql-functions-json.md) - [Spatial functions](https://www.hotdata.dev/docs/sql-functions-spatial.md) - [MCP Reference](https://www.hotdata.dev/docs/mcp.md) ## Integrations - [Python SDK](https://www.hotdata.dev/docs/python-sdk.md) - [Rust SDK](https://www.hotdata.dev/docs/rust-sdk.md) - [Agent Skills](https://www.hotdata.dev/docs/agent-skills.md) - [Ibis](https://www.hotdata.dev/docs/ibis.md) - [LangChain](https://www.hotdata.dev/docs/langchain.md) - [dlt](https://www.hotdata.dev/docs/dlthub.md) ## API Reference - [API Reference](https://www.hotdata.dev/docs/api-reference.md) - [Workspaces](https://www.hotdata.dev/docs/api-reference/workspaces.md) - [Query](https://www.hotdata.dev/docs/api-reference/query.md) - [Information Schema](https://www.hotdata.dev/docs/api-reference/information-schema.md) - [Results](https://www.hotdata.dev/docs/api-reference/results.md) - [Query Runs](https://www.hotdata.dev/docs/api-reference/query-runs.md) - [Uploads](https://www.hotdata.dev/docs/api-reference/uploads.md) - [Saved Queries](https://www.hotdata.dev/docs/api-reference/saved-queries.md) - [Indexes](https://www.hotdata.dev/docs/api-reference/indexes.md) - [Embedding Providers](https://www.hotdata.dev/docs/api-reference/embedding-providers.md) - [Jobs](https://www.hotdata.dev/docs/api-reference/jobs.md) - [Database context](https://www.hotdata.dev/docs/api-reference/database-context.md) - [Databases](https://www.hotdata.dev/docs/api-reference/databases.md) - [Usage](https://www.hotdata.dev/docs/api-reference/usage.md) ## Use cases - [Query every backend through one SQL surface](https://www.hotdata.dev/use-cases/unified-sql.md): One SQL path for connection-backed tables, managed databases, and SaaS sources registered in your workspace. - [Load parquet into managed databases and query it](https://www.hotdata.dev/use-cases/dataset-uploads.md): Load parquet uploads into managed-database tables that sit next to your connections in the same SQL namespace. - [Spin up ephemeral databases for scratch work](https://www.hotdata.dev/use-cases/sql-sandboxes.md): Throwaway databases for exploratory SQL and one-off loads, set to auto-expire so scratch work never lingers in shared environments. - [Run semantic search on embeddings from SQL](https://www.hotdata.dev/use-cases/vector-search.md): Vector indexes, semantic search from the CLI, and nearest-neighbor SQL in the same workspace as the rest of your queries. - [Upload and filter GeoParquet using spatial SQL](https://www.hotdata.dev/use-cases/spatial-parquet.md): GeoParquet as a managed-database table with distance, containment, and the rest of your SQL in one engine. - [Control spend and long-running hybrid queries](https://www.hotdata.dev/use-cases/query-spend.md): Async runs, stored results, connection refreshes, and job visibility without living in the terminal. - [Find slow SQL in history and add indexes for your filters](https://www.hotdata.dev/use-cases/index-tuning.md): Mine query history, inspect tables and columns, then add indexes that match real filters and joins. - [Publish database context for agents](https://www.hotdata.dev/use-cases/agent-context.md): Keep data model and team docs in a managed database so agents and the catalog stay aligned without pasting schemas into every prompt. - [Build shared context with your team](https://www.hotdata.dev/use-cases/shared-context.md): A shared data model plus optional team context, published through the database context API and read wherever queries run. - [Create full-text indexes and run keyword search from SQL](https://www.hotdata.dev/use-cases/text-search.md): BM25 indexes, keyword search from the CLI, and BM25-style SQL next to your relational filters. - [Explore tables and columns before you write SQL](https://www.hotdata.dev/use-cases/table-catalog.md): One CLI path to list connections, browse tables and columns, refresh after DDL, and page huge catalogs. ## Blog - [Benchmarking Concurrency on Hotdata, Snowflake and BigQuery](https://www.hotdata.dev/blog/benchmarking-concurrency-on-hotdata-snowflake-and-bigquery.md): For AI applications, concurrency is one of the most important performance metrics. Agents tend to make many requests at the same time, and the slowest requests impact the entire… - [ClickBench 10M: Hotdata vs Neon](https://www.hotdata.dev/blog/clickbench-10m-hotdata-vs-neon-no-indexes.md): We ran the ClickBench benchmark across 43 queries and 10 million rows (1.5 GB), comparing Hotdata to Neon. - [Bringing DuckLake to Apache DataFusion](https://www.hotdata.dev/blog/bringing-ducklake-to-apache-datafusion.md): We've just released a new version of DuckLake + DataFusion and have donated it to the datafusion-contrib repository. We’re delighted to bring a new lakehouse catalog format to the… - [How Hotdata accelerates Neon and Supabase](https://www.hotdata.dev/blog/how-hotdata-accelerates-neon-and-supabase.md): Agents rarely connect to a single database. Transactional databases like Postgres generally store core application data, but agents often need to query multiple systems and… - [Building Vector Search Into a SQL Query Engine](https://www.hotdata.dev/blog/building-vector-search-into-a-sql-query-engine.md): We added native ANN search to our DataFusion-based engine using USearch, Parquet, and SQLite — with adaptive filtering and got impressive results. - [Why Agents Need Version Control for Data](https://www.hotdata.dev/blog/why-agents-need-version-control-for-data.md): In the early days, developers could keep knowledge of the systems they worked on in their heads. You changed a file, ran a test, and moved on. That worked when codebases were… - [Agents Are Only as Good as the Data They Can JOIN](https://www.hotdata.dev/blog/agents-are-only-as-good-as-the-data-they-can-join.md): Now that everyone is building agents on diverse data sources and agents have better reasoning tools, how do you quickly assemble data from many different sources, in the right… - [Query engines for Agents](https://www.hotdata.dev/blog/query-engines-for-agents.md): We’re building a query engine, so I’ve been thinking about the difference between query engines built for humans and those specially made for agents. Regardless of the… - [Why we are building Hotdata](https://www.hotdata.dev/blog/why-we-are-building-hotdata.md): Over the past decade, data infrastructure prioritized throughput and scale over latency and interactivity, cementing large distributed warehouses as the default. ## Company - [About Us](https://www.hotdata.dev/about.md) - [Careers](https://www.hotdata.dev/careers.md) ## Legal - [Privacy Policy](https://www.hotdata.dev/privacy-policy.md) - [Terms of Service](https://www.hotdata.dev/terms-of-service.md)