mpurham

6 ms·

1.
by mpurham 3y ago
As for local LLM usage, that feature is coming very soon. This was in mind since day one as I'm a huge fan of various LLMs
2.
by mpurham 3y ago
Certainly, it's native, haha. I'm a developer who really dislikes Chrome-based apps due to their excessive memory consumption. The pricing in the app store is on the higher side because that version is quite outdated. However, the
3.
by mpurham 3y ago
When you hit the 'return' key in the app, it inserts a newline character. This design choice was implemented to facilitate the input of lengthy text.
4.
by mpurham 3y ago
thanks and there's currently no way to try it for free but, I've updated the pricing :)
5.
by mpurham 3y ago
1. Users retain control over their API key (hence fixed model pricing structure). 2. Upon initial installation of the application, it transmits specific information about the app, including its version, build number, and platform. It's
6.
by mpurham 3y ago
You're absolutely right – we're still in the early stages of the AI journey. Good luck with your AI tool as well
7.
Show HN: Olle – ChatGPT on Your Mac
12 points by mpurham 3y ago|15 comments
8.
Building a cross-platform developer tool as solo dev(marcell.me)
7 points by mpurham 4y ago|0 comments
9.
by mpurham 5y ago
Not sure how much coding you do nowadays but, I would highly recommend getting https://polypad.io/ license for quick macro tasks. I also recommend a https://www.masterclass.com/ subscription! No it is not p
10.
Show HN: Polypad – Scriptable Textpad for Developers (Cross Platform)(polypad.io)
2 points by mpurham 5y ago|0 comments
11.
by mpurham 5y ago
Definitely agree! Nowadays many developers skip over the design patterns, thinking abstractly, etc which over times causes technical debt. Many companies have very good processes in place and the above does not hold true for their employees
12.
by mpurham 6y ago
> The language doesn't matter much to me, I know enough of them to know that they are all very similar and thus easy enough to learn. The lie we all tell ourselves! yes it is true one can pick up a language fairly quickly but knowin
13.
by mpurham 6y ago
I remember spending hours sometimes trying to solve a single problem but I did not look up the solutions just worked my way through the problem sets. Algorithms definitely helped me prove code without explicitly writing it
14.
by mpurham 6y ago
I use to frown upon cross-platform js solutions until I built a cross platform app. While the initial app building part was no problem, packaging them for various platforms (macos, windows) was pure h*l.
15.
by mpurham 6y ago
It wasn't mentioned in the article but I'd say as developers, we often underestimate the amount of time a task/project would take. Learning how to properly estimate a given task and delivering it on or before your proposed ti
16.
by mpurham 6y ago
Users can care less about the app being native or non-native. I just want it to work and not slow me down. It's frustrating sometime when you switch machines and software you've purchased for a machine is not available across mult
17.
by mpurham 6y ago
The thing with programming in general is that everyone will write code differently depending on how they look at the problem. The best programming language is the one you are most familiar with. Yes you can write faster code in Ruby,Rust,Py
18.
by mpurham 6y ago
This was a great read and I definitely agree with the author that you have to be careful about data that you post online. This is a reason why I've created Polypad ( https://mattebot.co/polypad )
19.
by mpurham 6y ago
I'm building premium WordPress themes at https://radarthemes.com and occasionally writing at https://marcell.me
20.
by mpurham 6y ago
I faced this same decision a couple years back and one thing I would say is I have tried flat file services like blot ( https://blot.im ) and even building my own CMS and ultimately just ended up wasting time as I needed more func
21.
by mpurham 7y ago
I can closely relate to this post. Often times I build and develop software that is simply not worth my time. I think one important thing that you are doing well is capturing data for revenue. I think you should also take your hourly rate a
22.
by mpurham 7y ago
I was productive this year and was able to build many software apps from dashboards to video apps. I built: - Software at https://mattebot.co - Revamped https://marcell.me (native iOS, macOS development) - https:&#
23.
by mpurham 7y ago
Started writing macOS and iOS apps full-time and launched 4 apps for https://mattebot.co
24.
by mpurham 7y ago
You are estimating 2 weeks and I am assuming 40hrs/week (modest). Take your hourly rate and multiply it by 80hrs that is your estimated cost to build. In terms of finding if it is a needed solution, you will need to find your target cu
25.
by mpurham 7y ago
This!!! I have started many "side projects" with the notion that once I build it I will then find customers etc. This is terrible thinking and most of the times lead me to think my idea is horrible, when in reality I have done not
26.
by mpurham 7y ago
This is great! If this could somehow allow me to create UML diagrams I can see this becoming a leading tool.
27.
by mpurham 7y ago
I developed Semester Planner ( https://semesterplanner.com ) which is what helped me pay for college. It is a online student planner for college students that helps them keep tracks of their classes, notes, assignments, and docume
28.
by mpurham 7y ago
Awesome write-up and this post resonated with me as I have run saas companies in the past and while revenue was great it became a lot emailing customers, marketing, sales etc. I now am working on https://mattebot.co where I am
29.
by mpurham 7y ago
I tend to read blogs that cover PHP, C++, Linux and some iOS development here and there, etc. Here are some of my favs: - http://brendangregg.com - https://marcell.me - https://www.raywenderlich.com - ht
30.
by mpurham 7y ago
https://marcell.me I write about design and programming mostly iOS, C/C++, and Ruby.

More ›