bicx

12 ms·

151.
by bicx 2y ago
I try to generally keep up with the overall trends, but I’m an engineer at a resource-constrained startup, not a research scientist. I want to see real-world application, at least mid-term value, minimum lock-in, and strong supportability.
152.
by bicx 2y ago
That’s how I published my first webpage. Mindspring, baby! And then I later upgraded to a much more sophisticated web host with a cgi-bin.
153.
by bicx 2y ago
That's the thing though: a NIF execution isn't confined to the the BEAM process by its nature. From the Erlang docs: > As a NIF library is dynamically linked into the emulator process, this is the fastest way of calling C-code
154.
by bicx 2y ago
For Livebook, this looks really cool. Love that it calls CPython directly via C++ NIFS in Elixir and returns Elixir-native data structures. That's a lot cleaner than interacting with Python in Elixir via Ports, which is essentially exe
155.
by bicx 2y ago
Claude 3.5 has been fantastic in Windsurf. However, it does cost credits. DeepSeek V3 is now available in Windsurf at zero credit cost, which was a major shift for the company. Great to have variable options either way. I’d highly recommend
156.
by bicx 2y ago
I don’t have ADHD (that I know of) and still love Logseq. For me, it’s the perfect mix of notetaking, journaling, outlining, task tracking, and lightweight hierarchy/linking. I find that if I have to organize or categorize entries in a
157.
by bicx 2y ago
It's slower for sure, but capitalization does impart information: beginning of sentences, proper nouns, acronyms, and such. Sure, you could re-read the sentence until you figured all that out, but you are creating unnecessary hitches i
158.
by bicx 2y ago
The role of founding engineer is generally a rip-off if the salary isn't market rate. If you're not an actual founder, my advice is to ignore the equity entirely, try your best to get market rate, and look at the opportunity as a
159.
by bicx 2y ago
Why do you consider these something that can't be automatically detected?
160.
by bicx 2y ago
Honestly I doubt anyone can really answer with true authority on any meaningful timeframe beyond the next few months. What I'm more curious about is what other professionals consider the most resistant to automation and where to focus
161.
Ask HN: What are core engineering skills that will likely never be automated?
5 points by bicx 2y ago|9 comments
162.
by bicx 2y ago
Gotta scratch the schmootz off those little rollers too.
163.
by bicx 2y ago
New business idea: git diff, but it uses Mechanical Turk to hire an army of cross-eyed diff spotters.
164.
by bicx 2y ago
Is it possible that these may cause diseases that previously went unexplained?
165.
by bicx 2y ago
Coda is an incredibly powerful doc platform -- I first discovered it while working at a previous startup, and it became my absolute favorite. On my last week there, I built out a rip-off of Guitar Hero in Coda itself. We had a "multipl
166.
by bicx 2y ago
I'd like to announce my new `npm` package called `get-uuid`. Behind the scenes, it loads `everyuuid.com`, picks a random row number, and returns that UUID.
167.
by bicx 2y ago
Yeah, it's not too hard to trick it into generating some NSFW content ("mammary glands" for instance).
168.
by bicx 2y ago
Well Notion usually exceeds at least 3 req/s/core, so nothing to worry about there
169.
by bicx 2y ago
Well, it’s a published Notion site, and Notion is a powerful doc creation platform. It’s not really intended to be a performant publishing tool.
170.
by bicx 2y ago
True. Just not for the majority of the state at a usable level for the bulk of the population.
171.
by bicx 2y ago
I’ve come to the conclusion that California just doesn’t want anyone to live there. Once all the people are gone, it’ll be a green utopia once again.
172.
by bicx 2y ago
Factorio makes me realize that I don’t really like solving problems for the sake of solving problems (or optimization, for that matter). I like solving them for a tangible reason or because I I’m making someone’s life better. Problem solvin
173.
by bicx 2y ago
Trump just entered 0:08 instead of 8:00 on the grill timer. Rookie mistake.
174.
by bicx 2y ago
And finding human remains
175.
by bicx 2y ago
This kind of project has made me realize that somewhere along the way, I quit thinking of tech as a way to build anything fun. I need to rekindle that goofball spirit.
176.
by bicx 2y ago
I treat LLMs like a fallible being, the same way I treat humans. I don’t just trust output implicitly, and I accept help with tasks knowing I am taking a certain degree of risk. Mostly, my experience has been very positive with GPT-4o &#x2F
177.
by bicx 2y ago
Gemini is apparently hot garbage. I passed this input into ChatGPT (GPT-4o) and I got sensible answers.
178.
by bicx 2y ago
I understand your sentiment, but I also don't think it's fair to say people are being disingenuous. I don't work for an AI company, I just use it with decent results. My last project needed a way to tag certain types of busin
179.
by bicx 2y ago
Yeah this is a tired old narrative about SF. Remote-only companies can absolutely crush it and save money at the same time.
180.
by bicx 2y ago
Getting tired of seeing this concept of practically guaranteed hallucinations from any LLM used in production. I've used LLMs for various tasks, and if you tune your system correctly, it can be very reliable. It's just not always

More ›