Thunderduck (@thunderduckio) header
Thunderduck (@thunderduckio) profile picture

The AI-ready data engine. One SQL statement across Postgres, S3, BigQuery, Iceberg and many more — no pipelines, no warehouse project.

California
Joined October 2026
I just published Stop paying for BI seats: build a self-updating dashboard in Google Sheets. @thunderduckio medium.com/@yadiguzel/stop-p…
2
2
145
Which customers are you losing? A coffee shop asks Claude in plain English. Thunderduck joins its store DB with its marketing files: 3 of 4 Instagram-promo customers never came back. Referral? Only 1 in 5. No data engineer. No warehouse. 2-min demo 👇
1
1
39
Thunderduck now connects to @Cloudflare R2 and Basin ☁️🦆 Query R2 files and Basin Iceberg tables where they live, and join them in one SQL query with S3, Glue, Postgres, MySQL, MsSQL, Oracle, Postgres,ClickHouse and BigQuery No copies. No ETL to babysit. thunderduck.ai/blog/cloudfla…
1
1
1
26
A query searched ~1B rows to return 886. 438 s, CPU at 10%. Rewriting it wasn't an option, so we tuned DuckDB instead: more threads than cores, memory headroom, a higher join-key pushdown limit, sorted tables. 438 s → 101 s. Same machine, same SQL. thunderduck.ai/blog/billion-…
12
Thunderduck retweeted
Welcome Russia! #VPN
1
3
126
Hello, world. 👋 Making your data AI-ready shouldn't take two quarters and a warehouse migration. Thunderduck queries your data where it already lives, with one MCP endpoint for every AI tool. DuckDB does the heavy lifting. The thunder is ours. ⚡🦆 thunderduck.ai/blog/hello-wo…
2
11