phamilton

5 ms·

1.
by phamilton 2mo ago
Any graphql library would work. I like async-graphql except compilation is a bear due to all the macros. We wired LISTEN up to subscriptions using tokio mpmc channels. It was a great moment of "fearless concurrency", it all just w
2.
by phamilton 2mo ago
Just sharing a data point and experience. We had a lot of success with LISTEN/NOTIFY when we paired it with a Rust graphql subscription broker. 10s of thousands of subscriptions, but only 3 or 4 LISTEN connections (one for each host).
3.
by phamilton 3mo ago
Just a set of things too minor to move off of it but annoying enough to not want to start with it. My list: No `explain (analyze,buffers)`. Instant DDL has some warts (e.g. fk, metadata locks). Query planning bugs (actually... query plannin
4.
by phamilton 3mo ago
I've always known how to do zero downtime migrations. The question has always been "Is the engineering cost worth it?" > It only makes backwards compatible deployments easier insofar you're able to do the overhead for
5.
by phamilton 3mo ago
> Old things are new again. This is exactly how I've felt. I've read some many old papers and books and found great techniques that are even more applicable than ever.
6.
by phamilton 3mo ago
> perfect refactor-proof testing contracts are still usually pretty hard to design Not sure about hard, but definitely rare and we as an industry are under-skilled in these areas. We have decades of research and tools for testing and ver
7.
by phamilton 3mo ago
That's not at all what I meant. I mean "We can't build X because our code structure makes that difficult" has an opportunity cost of the value of X. I don't think the future of dev work is being a bureaucrat. I&#x27
8.
by phamilton 3mo ago
I'm not talking about time. I'm talking about safety. The amount of times I've seen "I refactored it, but I'm not confident enough to take it to prod" is significant. Being able to go faster but still not ship
9.
by phamilton 3mo ago
The cost of restructuring has also gone down. The cost of shoring up behavior with tests ahead of a restructure has gone down because of AI. The cost of implementing a zero downtime migration has gone down because of AI. A big part of the r
10.
by phamilton 3mo ago
MTP on a MoE is hit or miss. If you're bottlenecked on memory, MTP can increase the number of active experts (like any batch processing would), which can eat away gains from it.
11.
by phamilton 3mo ago
Generation is basically just memory bandwidth math. Each token has to read all the active weights. I think that's around 40B parameters active. At a 4-bit quant that's 20GB. With 100GB/s (replace with whatever your bandwidth
12.
by phamilton 4mo ago
My favorite lens on SQLite is that it is actually two things: 1. A robust durability implementation 2. A library of high performance data structure and algorithms The fact this it's SQL is nice, but those two attributes are what make i
13.
by phamilton 4mo ago
I wonder how Sonnet vs Opus stacks up in a similar time-based comparison. How far behind are open models compared to Sonnet? It may be that the absolute SOTA models are way ahead of open models, but the gap in the mid tier really does feel
14.
by phamilton 4mo ago
I'm running opencode with qwen3.6-35b-a3b at a 3-bit quant. I also have qwen3.5-0.8b used for context compaction. I run with 128k context. It's usable. I set it loose on the postgres codebase, told it to find or build a performanc
15.
by phamilton 4mo ago
I have some experience in this. Reach out (email in my bio) I would love to chat.
16.
by phamilton 5mo ago
Long agent runs make such a difference. We focus a lot on new models and long context, but the bigger impact is in automatic verification. I've been leaning in more on e2e test suites. They are slow, brittle and inefficient. But that&#
17.
by phamilton 5mo ago
Gemma4, qwen3.6, deepseek v4, mimo, glm 5/5.1 all do MTP.
18.
by phamilton 5mo ago
From the same org as pgrx is https://github.com/pgcentralfoundation/plrust , which _is_ supported on RDS (but not Aurora). Since it's a procedural language, you can't do things like create a new index implemen
19.
by phamilton 5mo ago
I've definitely done some vibe-coding with the explicit intent to reduce memory usage.
20.
by phamilton 5mo ago
The announcement was that it secured $50M in financing, sold the shoes business for $39M leaving $20M or so in cash. An empty public company with $70M in financing to enter a hyped market was valued at $115M. The stated intent is to spend t
21.
by phamilton 5mo ago
Given the premise that zero day exploits are going to be frequent going forward, I feel like there is a new standard for secure deployment. Namely, all remote access (including serving http) must managed by a major player big enough to be p
22.
by phamilton 5mo ago
Isn't this just a SPAC? The shoe business was sold, a shell of a public company was left, and it essentially acquired a brand new company focused on AI.
23.
by phamilton 5mo ago
I just got it last week. Still a few quirks, but positive so far.
24.
by phamilton 5mo ago
I dropped my MBA on concrete and the edges got dinged up and sharp. A bit of 220 grit sandpaper and all the sharp edges are smooth and it actually looks pretty cool. I was grimacing at first but now I like the feel.
25.
by phamilton 5mo ago
More like Fire Emblem
26.
by phamilton 6mo ago
15 years ago I was an intern at Micron and learned they passed on a contract with Apple because Apple insisted on discounts and there wasn't a compelling reason to reduce profit at Micron. So yeah, Apple probably does pay less. But the
27.
by phamilton 7mo ago
> For those building with a mix of bash and custom tools, Gemini 3.1 Pro Preview comes with a separate endpoint available via the API called gemini-3.1-pro-preview-customtools. This endpoint is better at prioritizing your custom tools (f
28.
by phamilton 7mo ago
> Several dollars per million tokens The flagship, glm-5, is $1/M input tokens. glm-4.7 is $0.60/M input tokens.
29.
by phamilton 7mo ago
Not yet. But what's the actual goal here? It's not to have a native Wave Race 64. It's to improve my intuition around what sort of tasks can be worked on 24/7 without supervision. I have a hypothesis that I can verify th
30.
by phamilton 7mo ago
Step 1: Decompile into C that can be recompiled into a working ROM. In theory, it could be compiled into the same ROM that we started with. Consistent ROM hash is the main success criteria for the OoT decompilation project. Have it grind un

More ›