621–630 of 922 posts
Right now we're trying to figure out our funnels. We've had some pretty good success doing things that don't scale particularly well, like going to immigration events. Now I'm working on getting better visibility across the country (SEO, Google Ads, etc...).
I'm learning a lot, but it's slow going.
These are audit data like external system request/responses for possible investigations. This saves a lot of space. Initially written in Python, now practicing with Rust. Container images is 2.2MB small :)
I'm several months into https://www.clearboxlegal.com , a service for affordable legal help with citizenship applications. Right now we're trying to figure out our funnels. We've had some pretty good success doing things that don't scale particularly well, like going to immigration events. Now I'm working on getting better visibility across the country (SEO, Google Ads, etc...). I'm learning a lot, but it's slow goin…
I was laid off a month ago so my biggest project right now is finding work.
I live next to a school, so there's a low speed limit (30 km/h). Still, people drive like race drivers and the city hasn't ever responded to the residents' hopes of introducing a speed camera. I wanted to have some data on how many people speed, the max speed recorded, that sort of thing. Things the city should be doing after many complaints of dangerous driving and people being almost killed on zebra crossings. I ha…
The guy at Not Just Bikes will tell you that enforcement will never work nor happen and that the only way to get people to slow down is to design the road so it doesn't feel safe to drive fast. The road next to my house has a speed limit of 20mph but most cars go 45mph because it's a straight road 4 lanes but space for 6. No bumps, no curves, wide. Effectively it feels like you should be driving fast. If I go the spe…
Our neighboring town has a beautiful system in place. It tracks your average speed over +- 100m. If you speeded, you get a fine and even better: the light always turns red and you have to wait for a minute. If you run that red light, you get another (heftier) fine. If you drive normally, the light is always green.
It's beautiful. People actually drive slowly and safely there. And good, because that system is installed perfectly around the school and doesn't hinder anyone else. If they did this anywhere else you'd argue it's a moneygrab, but this is imo a great way to actually get people to drive slow in a school zone.
In contrast the only thing our town has done is install a bunch of big concrete flower boxes (?) that hinder your vision and causes conflicts on the road. The idea being that you'd have to slow down and that you're no longer on a straight wide road. It's absolutely only made things worse. You need to give way to others but you literally can't see them, or you need to pull weird manoeuvers to let them pass. And in the meantime cyclists need to swerve around them or are hidden behind those boxes. It's terrible.
https://github.com/btahir/open-deep-research
An open source alternative to Gemini/OpenAI deep research. I'm experimenting with this cool flow chart workflow that I think can be a cool twist on generating reports for deep research.
Check it out and lemme know what you think. :)
https://github.com/turbolytics/sql-flow
The goal is to create a stream processing framework that supports SQL jobs. Apache Flink supports this but is very heavy-weight overall. I work with cost constrained companies that just can't run Flink but still want access to high performance streaming primitives.
We are taking a slightly different approach from other competitors by building cloud-native tools for engineers. SQLFlow is built on DuckDB, Native Kafka Library, and Arrow. This allows SQLFlow to handle ~70k+ events / second with low memory overhead (~250MiB).
Would love your questions, thoughts and feedback or feature Ideas! Thank you
Don’t have a background in ML, so mostly just for learning purposes
Been playing a lot with the MNIST dataset. Trying things like training only on the examples the model gets wrong, or training only on random samples of the image (ie. using only a small subset of the pixels of the image as the input), or creating one model per label to overfit the data and then merging the models for testing, or just testing performance of different architectures and frameworks on the same problem
Earlier quoted context omitted.
i think part of the problem is 30mph/48kmph doesnt feel fast - and to get people to drive slower manufacturers need to design cars so 20mph/32kmph feels faster
There is another way: make the road feel fast. Thankfully it doesn't need to be via bad road surfaces or horrible things like speed bumps that only encourage boy racers and reward large vehicles, making the road narrower with high kerbs or other physical obstacles force drivers to drive slowly and pay attention, otherwise they'll physically damage their vehicle. In a way it evens the playing field, currently cars can…
Honestly it's already pretty narrow and crowded here. I'm a regular driver here (duh) and it honestly feels dumb to drive faster than 30 because a cyclist can pop out of every corner. It's crazy to me that people want to, and do drive faster than that. But here we are. People suck.