When Columbia's @DAP__Lab publishes BranchBench comparing your product to competitors, you pay attention. Read our full breakdown and response here: https://t.co/SB2UsoT8ZT
DumboDB, a version-controlled document DB, uses Dolt's engine and now adds garbage collection!
Neil's blog dives in π https://t.co/wXT2Mnquxw
Start versioning π· https://t.co/K41bpBGqIL
π Doltgres now supports functional indexes!
You can now index expressions and function results in our version-controlled Postgres database.
β‘οΈ Tested a 37x speedup on 300k rows from 150ms down to 4ms.
Try it safely on an isolated branch! https://t.co/m0rt0xX5T3
Beads gives coding agents like Claude Code persistent memory with Dolt's version-controlled infrastructure. Start in classic mode, then evolve for concurrency. Learn how to level up here: https://t.co/gN1xikt4Ga
We used Claude Code to migrate DoltHubβs docs from GitBook to Astro.
Great at tests, redirects, link cleanup, Markdown, and source-grounded fixes.
Still needed human eyes for anything visual.
Full post: https://t.co/RskVF4m8md
See the update: https://t.co/L6JrnzLIO3
What do engineers do with database merge conflicts? Timothy Sehn shares 3 Dolt-native patterns. https://t.co/yT9aQeONoU #Dolt#MergeConflicts#GitForData
@kylegawley 2025: The database was the bottleneck
2026: AI writes to databases.
2027: Turns out data needed version control, too.
We've been thinking about this for a while. https://t.co/QVBZj2PA0Z
Dolt 2.0 speeds up table scans 40% vs MySQL by storing small TEXT, JSON, GEOMETRY, and BLOB values inline. Read about the improvements below https://t.co/tiGEndhSeE
DumboDB is Dolthub's new version-controlled document database, built on Dolt's storage engine.
Performance results are in. Neil Macneale ran DumboDB 0.1 head-to-head against MongoDB.
See how it holds up π https://t.co/BOgRrnbMR8
Every Dolt blog post ships with a header image. We just put it in motion. https://t.co/Cpj4lS2P2q
No AI was used in the making of this video.
Trad art approved.
TPC-C is the gold standard for database performance testing, simulating real-world transactions such as orders, payments, and deliveries.
The blog breaks down how TPC-C works and what we're still working toward.
π https://t.co/enY4rLH62t
Tim Sehn used Codex (GPT-5.5) to cut DoltLite comments from 9,359 to 193, boosting agent performance. #doltliteπͺΆ #DatabaseForAI https://t.co/Y5zGX3Q4XG https://t.co/B1wILxbZrg
We just released DumboDB, a NoSQL DB from DoltHub using MongoDB tools, with unique indexing and version control.
Read what's under the hood: https://t.co/XQN3Sor8h7
π Dolt 2.0: The first version-controlled SQL database outperforms MySQL with 13% faster writes, 5% faster reads, 30-50% storage savings, auto garbage collection, and version-controlled vectors. https://t.co/UpT8ljHJGI #OpenSource
Can we interest you in a version-controlled document database? We shipped DumboDB 0.1, a MongoDB clone with Git-style version control. GitHub: https://t.co/K41bpBFSTd Full writeup: https://t.co/1hGdYcoF68 Come shout your requests on Discord: https://t.co/B1wILxbZrg
Traditional DB: "Your AI agent modified 47,294 rows"
Dolt:
β `dolt diff` (see exactly what changed)
β `dolt blame` (which agent did it)
β `dolt revert` (undo the disaster)
Version control, not amnesia.
#GitForData#DatabaseForAI