Reddit’s internal media processor is incredibly rigid and basically throws a silent tantrum if a video isn't perfectly formatted. It's so picky and won't tell you how to fix it.
Are there any other conversion tool options aside from this one?
Several of my coworkers and I were laid off a year ago from remote jobs (I was working as a project manager at a San Diego SaaS company that was bought by private equity). A few of us were lucky to barely survive a round of layoffs the year prior, but after we trained a couple teams in the Philippines and India to do our work, they finally closed down our teams stateside.
With all the layoffs we keep hearing about in tech, how are you all faring? Have you had to pivot to different types of roles? Any advice?
Edit: fixed typo. Philippines, not Philippians
I honestly have no idea why you'd want this, but it was fun to experiment. I believe this may be the best pure CSS cursor tracking that exists. I'm happy to be proven wrong.
https://codepen.io/Andrew-Fisher-the-decoder/full/GgraMzd
Absolutely no JavaScript used, just lots and lots of wasted time.
I know tools like this exist, but they seem to be over-complicated and limited. I tried to keep mine easy for even me to use, as I am not a font designer. It also does not use AI, so there are no limits (you can use your own LLM to generate the font image if you want, and the tutorial will walk you through that and provide the prompt, but you don't have to do that).
Please try it out and give me feedback here and I will do my best to add any requested updates.
Edit: Why I built this. This conversation here led me to believe there must not already be a satisfying option for creating your own font. I initially built the tool for me, to see how possible it was to create a font, and I was surprised that I was able to create my own font pretty easily. Your results may vary.
I've been fascinated lately with the original version of the song Bang Bang, which I was surprised to learn was written by Sonny Bono and recorded originally by Cher in 1966.
On the surface, the song appears to be a simple, playful, nostalgic, reminiscing of a lost lover. Nancy Sinatra's version certainly reads like that. However, when you listen to the lyrics, and especially when you watch Cher's music video, something seems off. There are religious overtones, splashes of color and images of poppies, and an ending that implies death. The style of the arrangement in the original is also a strange departure from what you'd expect from a song about children falling in love and breaking each other's hearts. It sounds like it's inspired by gypsy jazz. It was arranged by Harold Battiste, an admirer of Charlie Parker.
I started thinking about the term "horses made of sticks" and realized the phrase could easily be referring to the act of shooting up heroin, as both "sticks" and "horse" were drug terminology common from the 1940s and through the 1960s. The term "bang' is also slang that refers to shooting up heroin. Obviously, "shot" also fits.
But "I was five and he was six" is where I turned to jazz. Here me out:
There were two common jazz group sizes in the 1940s - the quartet and the quintet. Charlie Parker was among the very first to record under a sextet lineup in the late 1940s. Could 5 and 6 be referring to this shift? It would also explain that "awful sound" inspired by heroin, a frenetic bebop that strayed from the jazz of the time.
"He wore black and I wore white" could refer to the different styles of jazz that appealed to black audiences and white audiences.
I could go deeper, but honestly all of this is conjecture, and I don't know that much about jazz history. I'll just end with who I think the song is about. My guess, from what I can find, is that it's written from the perspective of Miles Davis about his relationship with Charlie Parker.
I'd love the perspective of more educated folks here.
Use free alternatives wherever possible. For example, Google's AI Mode is a great model for many use cases, and it's conveniently and freely available right from Google.com. Google's NotebookLM (notebooklm.google.com) is also free, and a fantastic tool for summarizing hundreds of sources. It's essentially a mini-custom model, without the overhead. Many chat-style models have free tiers that you can cycle through for researching a certain technology, debugging an error, or even coding a simple MVP.
Always build small first. When coding, always build out a small MVP first in a cheaper model, like ChatGPT or Gemini. Even if you already have an app in progress, for any new major functionality you want to add to your app, first build it independently, separately, on a small scale, using a cheap LLM. You can upload a screenshot of your existing app, or paste some of the existing code to start the conversation. If needed, you can use free, online CDN libraries and APIs to pull in functionality. Have the LLM write the MVP in JavaScript, as that is natively supported in all browsers and easily translatable into your codebase once the MVP is tested and working. When it's ready, you can copy the working code into your ongoing conversation for your main code base, along with the prompt: "Port this design/functionality over to the existing application" (include any specifics to ensure the correct pieces are integrated). Note: If you're wondering how to test things without integrating your database, you needn't worry. You'll be surprised what can be done entirely in JavaScript, without a backend. For example, check out the many free web tools available on https://getmoredonefast.com, and also check out https://noisefixer.com. These are built with no database, and run entirely in the browser.
Tell your LLM to use "code golf" techniques. Code golf is a competition where coders attempt to build a functioning app in as little code as possible. Typically, this is not the correct way to write code, because it's often not standardized or scalable. However, LLMs tend to be very good at balancing efficient coding with best practice, as long as you explicitly tell it to. If you don't, an LLM will gladly write as much long, verbose code as it can, running you out of tokens and exploding its context beyond its capability. It will use up your budget and then start forgetting what it wrote earlier. Don't let that happen. I often start each conversation with "Code golf an application that..." And I also like to instruct it to use "efficient, infinitely scalable architecture" and "optimized" code. In other words, just make clear that brevity is a priority.
If you are coding an app that itself will use AI, see if you can build the app without integrating the AI. This will save you a lot of money. I often see people build a new AI-enabled app that ends up costing them thousands in unexpected fees that they easily could have avoided by building it differently or building something else entirely. You can make an app "smart" without AI by using some tricks like deterministic decision trees, free APIs, or a database lookup table. Don't get swept up in the trend of adding AI to something that doesn't need it. Challenge yourself to build it without integrating AI, and you will get further than you realized you could, and you'll likely be saving yourself money in the end.
Try to build in a way that is optimized for your server pricing scale. Some hosting platforms charge based on total bandwidth, and some charge based on number of queries. The limits will also be different for other variables like users, quantity of database tables, and any number of other things. You can integrate third party platforms to manage users, security, and file uploads, and you can use APIs for some functionality. Shop around to find the cheapest options for your use case, and have your LLM write your application in a way that takes the best advantage of your setup. Share the pricing tables in your conversation, along with a prompt like "Build this in a way that will keep the app infinitely scalable and minimize costs."
Now that you're thinking about all the above, you'll probably find other ways to optimize and maximize efficiency. Feel free to share them below.