jtmarmon

I'm hustling. I'm playing the game. I'm doing what it takes to make this thing mainstream and a viable, practical choice for individuals and companies. If you talk shit about Zig in public, I'm going to fight back. But I respect you. I see you. I understand you. I don't hate you. I would literally buy you a drink.

Andrew Kelley - An Open Letter to Everyone I've Butted Heads With (2025 Aug 29) https://andrewkelley.me/post/open-letter-everyone-butted-hea...

I've had the same problem for a couple years - specifically the GPU crashing. Had a very hard time isolating the issue - seems like a mix of static + the EMI spike OP talks about (it happens most reliably when I stand up quickly from my desk chair).

My guess is that, like OP, we're both getting interference in the our DP connections, and that that interference is in our cases causing the GPUs to crash.

Haven't had a chance to try ferrite cores yet but that was going to be my first test.

Curious what system specs you have in case we have overlap in anything that could isolate the issue. Mine: https://pcpartpicker.com/list/Xpdb8Z

My guess is that, in the vein of the "monitor flickering" symptom, your PC sees the DP/HDMI cable disconnect and reconnect due to the static. The reconnection wakes it up.

I pretty much only use o1 for more rigorous tasks (coding/math). Eg the other day I gave it the tax rates for my area and used it to explore different tax scenarios

I’m no expert on docker but I thought the hashes for each layer would already be computed if your image is built

Great writeup. To add an example, I personally use JSON for most of my work, but have found myself using XML for certain AI use cases that require annotating an original text.

For example, if I wanted an AI to help me highlight to the user where in a body of text I mentioned AI, I might have it return something like:

<text>Great writeup. To add an example, I personally use JSON for most of my work, but have found myself using XML for certain <ai-mention>AI</ai-mention> use cases that require annotating an original text with segments.</text>

SelectIQ (YC W23) | Founding Software Engineer (Fullstack) | NYC or REMOTE (US) | Full-Time | https://selectiq.ai/

SelectIQ works with clinical trial sites and their referring physicians to automate the patient referral process, massively speeding up enrollment.

Bringing new drugs to market is primarily rate limited on the speed of clinical trials, which are themselves bottlenecked by patient recruitment. 80% of clinical trials don’t meet their recruitment timeline, and being behind on recruitment can cost pharma companies millions of dollars per day.

We're looking to bring on our first engineering hire to help us build out the platform and build an amazing culture. More details and application can be found here: https://www.ycombinator.com/companies/selectiq/jobs/4aRBMcX-...

Feel free to email me directly - jason@selectiq.ai

I recently evaluated Dagster, Prefect, and Flyte for a data pipeliney workflow and ended up going with Temporal.

The shared feature between Temporal and those three is the workflow orchestration piece. All 3 can manage a dependency graph of jobs, handle retries, start from checkpoints, etc.

At a high level the big reason they’re different is Temporal is entirely focused on the orchestration piece, and the others are much more focused on the data piece, which comes out in a lot of the different features. Temporal has SDKs in most languages, and has a queuing system that allows you to run different workflows or even activities (tasks within a workflow) in different workers, manage concurrency, etc. You can write a parent workflow that orchestrates sub-workflows that could live in 5 other services. It’s just really composable and fits much more nicely into the critical path of your app.

Prefect is probably the closest of your list to temporal, in that it’s less opinionated than others about the workflows being “data oriented”, but it’s still only in python, and it deosn't have queueing. In short this means that your workflows are kinda supposed to run in one box running python somewhere. Temporal will let you define a 10 part workflow where two parts run on a python service running with a GPU, and the remaining parts are running in the same node.js process as your main server.

Dagster’s feature set is even more focused on data-workflows, as your workflows are meant to produce data “assets” which can be materialized/cached, etc.

They’re pretty much all designed for a data engineering team to manage many individual pipelines that are external from your application code, whereas temporal is designed to be a system that manages workflow complexity for code that (more often) runs in your application.

This is a pretty good way to go bankrupt when you have a real health emergency

EDIT: I doubt OP is reading this but I would highly recommend they get insurance before giving birth...sure, they won't cover your at home birth, but they will cover your hospital stay if you end up having potentially fatal complications

If you like startups and hard tech, check out Jason Carman's S3 channel - https://www.youtube.com/@jasoncarman/videos

Slow build times make everything slower, including refactoring tech debt (which means people are less likely to do it).

Congrats on the launch, looks like a great product.

Out of curiosity - the demo on the site suggests the local first latency is like 10-40 ms. I would think reading/writing to SQLite locally would be on the order of 1ms or less. Why is that?

I just had it fix some outdated copy in a part of the UI. The nice thing is I didn't have to find the file myself, I just described what was wrong like I would a junior eng and let it find and fix it. Worked on the first try!

Just merged my first simple PR with sweep. This is going to be so useful for the kind of things that would take 5 minutes to do but get procrastinated for weeks because you just can't find the time to context switch for it.

Congrats on the launch!

Obviously it is the argument, that's why it's titled "Tailwind and the Femininity of CSS". Or do you think the 5 paragraphs about how sexism causes people to dislike CSS is a non sequitur from the introduction about how the author prefers Tailwind to CSS. Break it down for me!

"It has nothing to do with gender" is with regard to the ratio of BE to FE people in engineering leadership. I'm not making the argument that gender bias plays no role in promotion anywhere. It is possible for some things in the world to not be explained by gender bias.

This article is hilariously bad. The argument goes:

I like CSS more than Tailwind -> Why don't people like CSS more? -> Maybe because 'CSS, which makes things look ‘pretty’, is considered feminine'

You're entitled to like CSS more, and I could even agree making things look pretty is feminine coded, but it obviously doesn't explain people's preference for Tailwind because Tailwind also exists to make things look pretty.

---

I've worked with plenty of female engineering leaders, and most of them have a backend background.

The reason for this imbalance has nothing to do with gender, but entirely to do with criticality. Given that frontends tend to read/write from the backend, the domain model is usually owned by the backend in most apps, meaning that capability design and expansion is gated by the backend.

Not to mention that screwing up your backend architecture is in 95% of cases a much much deeper problem than screwing up your frontend. A data migration is basically always harder than redesigning the UI for some app.

This looks great! Is there a reason one couldn't use this with v4 UUIDs? A quick test shows that they encode/decode just fine. Wondering if I could use the encoded form as a way to niceify our URLs without having to change how the IDs (currently v4 uuids) are stored

1. Airpod pros have spatial audio

2. Their presentation showed someone using it with airpods on a plane

https://twitter.com/techAU/status/1665790510093697024

z...z.z...zz..z...z.z...z..z.fz...z.z...z.z....a..ba..ba..ba..ba..ba..ba..ba..ba..ba..ba..ba..ba..ba..ba..ba..ba..ba..ba..ba..ba..ba..ba..ba..ba..ba..ba..baz.z

Would be really cool to see you guys become a managed service on Fly.io - see what they did with Upstash & Redis:

https://community.fly.io/t/preview-managed-upstash-redis-wit...

They don't have a managed postgres offering yet ("this is not a managed postgres" - https://fly.io/docs/postgres/getting-started/what-you-should...)

Except my app isn't down due to an app error but a failed host in EWR which I couldn't escape from (due to the concurrent scheduling issues) https://status.flyio.net/incidents/v2dshzvy1mcl

EDIT: recognize that these may be poorly timed but unrelated incidents, but it has been frustrating to be trapped on a broken box for 12 hours and have the status page telling me it's just new deploys that are borked :)

I think this status page is inaccurate - hosting is affected. My app is _unfixbly_ broken right now. I have an app on fly whose VM appears to have died due to these issues, and because deploys and restarts are broken, I have literally no way of fixing it. https://community.fly.io/t/ewr-app-is-completely-inaccessibl...

I found myself wanting this recently (on the consumer side) so really excited to see this!

edit: One suggestion - maybe have a discord server that people provide support through, that way there's already an (IM-based) comm channel setup. Would def reduce the friction for me (and increase the chances I become a repeat customer)

You usually hear it in the context of someone who's had a long enough career to forget how hard "just writing code" is when you're a noob ;)

In addition to a built in log drain, my other FR would be better tooling for specifically build and deployment logs. When upgrades or remote builds fail, I've found finding the logs for either of these to be a bit of a challenge. Heroku has that nice deployment logs interface that I'd love to see in fly.

To double tap the sibling's replies here - I'm a fly user and found it challenging to locate the release_command docs. Highlighting these in the docs somehow would be a good idea since I think _most_ apps will need something like this.

See a tech stack that seems cool/sexy/interesting? Make something up to build with it over a weekend to try it out. Put your blog on Kubernetes just for fun.

The article alludes to laser comms - NASA is developing[0] laser based comms systems (as opposed to radio frequency) which would allow gigabit speed downloading of data. Hardening this technology to send back more raw data is probably a lot more straight forward than trying to do image processing on board.

[0] https://www.nasa.gov/mission_pages/tdm/lcrd/index.html

Should've built it in elixir, I heard it's webscale

This list of items is called debugging, and it's probably your job and not the non technical user who reported it to you ;)