mscccc Profile Banner
Mike Coutermarsh Profile
Mike Coutermarsh

@mscccc

Followers
11K
Following
166K
Media
1K
Statuses
19K

Building really fucking fast MySQL and Postgres databases @PlanetScale

Pennsylvania, USA
Joined December 2009
Don't wanna be here? Send us removal request.
@p_reynolds
Patrick Reynolds
7 hours
PSA: if you use Claude Code and @planetscale and have the PlanetScale CLI installed, try this: claude mcp add --transport stdio -- planetscale pscale mcp server Then you can ask Claude Code questions about your databases.
1
3
10
@jsneedles
Jeff Needles
7 hours
Took @mscccc’s new header photo
1
1
7
@matsonj
Jacob Matson
13 hours
so i've been fooling around with planetscale for the last few weeks, and as a guy who spend a lot of time working on apps backed by sql server, its been interesting - its thread time!
3
5
104
@mscccc
Mike Coutermarsh
17 hours
Notice how most of the replies and quote tweets are near identical AI slop about microservices.
@SumitM_X
SumitM
21 hours
As a developer, have you asked yourself - why do we need UUIDs when auto-increment IDs exist?
7
1
35
@fatih
Fatih Arslan
1 day
Smaller Metal plans are now live. Definitely check it out if you wanted to use our high-performance databases.
@PlanetScale
PlanetScale
2 days
$50 PlanetScale Metal is live. https://t.co/s5lMfJijMu
0
1
17
@mscccc
Mike Coutermarsh
2 days
6 levels of OLTP database query latency: Nice: < 1ms Acceptable: 1-3ms Probably fine: 3-9ms Slow: 10-20ms Concerningly slow: 20-100ms Fix it now: 100ms+
13
18
322
@mscccc
Mike Coutermarsh
4 days
Just realizing now why the planetscale site is so easy to work with in cursor. We’re no CMS and markdown only as well. Having the context of every doc, blog, page in a repo is very llm friendly
@leerob
Lee Robinson
5 days
I migrated cursor​.com from a CMS to raw code and Markdown. I had estimated it would take a few weeks, but was able to finish the migration in three days with $260 in tokens and hundreds of agents. Here's how I did it + all my my usage stats. https://t.co/QIAOmLsffx
4
1
87
@mscccc
Mike Coutermarsh
4 days
App spending 1500ms talking to the database and 30ms rendering json. “If they used a modern framework it would be faster”
16
9
278
@mscccc
Mike Coutermarsh
4 days
I worked there 5 years ago, which means I don’t know what the current challenges are or how to fix them. But I’m very confident that it’s so complex it cannot be blamed on any specific framework or technology.
2
0
27
@mscccc
Mike Coutermarsh
4 days
As a person who worked on GitHub dot com and now builds PlanetScale using Rails + React. The answer to this question is: No
@kirso_
Kirill So
5 days
@jaredpalmer @thdxr so is rails basically a bottleneck when coupled with modern frontend?
15
4
257
@jsneedles
Jeff Needles
6 days
Another benefit of leaving San Francisco. Real entertainment available.
@acoustik
Ajay Kulkarni
6 days
Only in SF can you watch a panel of MCP pioneers, including the person who first created it at Anthropic, talk in front of an audience of 300 people... on a random Wednesday night.
2
1
17
@mscccc
Mike Coutermarsh
7 days
My "favorite" one of these was someone from Google joined GitHub and wrote a post a couple weeks in about how GitHub used google docs wrong (we used GitHub for everything... obviously).
@corywilkerson
Cory Wilkerson
7 days
There is absolutely no worse move than joining a company and sharing your written assessment of everything that’s wrong w/ five total days of tenure. 100x screw up if posted on an internal blog like a thinkpiece.
2
3
293
@nikunj
Nikunj Kothari
10 days
Ok it's getting truly banana town in VC and startup land. And, I think the people who will get HURT the most in this might be the early employees. So here are some things you must absolutely know and questions for you to ask as you enter job hunting season in January.. 1)
Tweet card summary image
pitch.com
Easily build and share beautiful presentations with Pitch. Start for free on pitch.com
25
24
311
@hollylawly
Holly Guevara
11 days
Many such cases
@brankopetric00
Branko
12 days
"Our database won't scale". Database: - 40GB total data - 12 queries per second - 0 indexes on query columns - N+1 queries everywhere - 200ms average query time Solution: - Shard across 12 databases - Add read replicas - Implement caching layer - Switch to "web scale" NoSQL
1
4
60
@mscccc
Mike Coutermarsh
11 days
we would never do postgres
@PlanetScale
PlanetScale
11 days
From the archives of our internal Postgres hack week this past spring.
2
0
13
@mscccc
Mike Coutermarsh
12 days
1.5x better performance at half the cost 🥵
@PlanetScale
PlanetScale
12 days
Core flow AI's move from Supabase to PlanetScale led to 1.5x better performance at half the cost. https://t.co/GXCsYyvclg
2
1
30
@mscccc
Mike Coutermarsh
12 days
And I remember we were worried about how this would "scale" or be annoying. How can we auto generate it? Will it break? Should we use graphql? Anyway, almost 5 years later, it's never been an issue. No one talks about it. File is still there and people still update it.
4
0
24
@mscccc
Mike Coutermarsh
12 days
One interesting thing about working somewhere for a while. You get to see how tech decisions that felt big at the time work out. For example. We have a react frontend that talks to a Rails API. Early on, we created a types file and manually updated it for each Rails API endpoint
1
0
32
@mscccc
Mike Coutermarsh
15 days
Millennial code review feedback
2
0
39
@simeonGriggs
simeonGriggs
16 days
👋 Day 1 at @PlanetScale
24
4
93