Hey everyone,

I’m a high school developer looking to build a tailored application for a technical project. I want to build a solution for a genuine, everyday headache that real people or niche hobbies/industries deal with, rather than making another generic dashboard.

My goal is to build a software suite that combines:

  1. Predictive Machine Learning / Classification (e.g., predicting trends, risk scoring, categorising complex inputs)

  2. Custom Algorithmic Logic (e.g., route optimisation, spatial pathfinding, data compression, audio/signal processing)

  3. Data Management & Analytics (database logging, parameter tuning, clean user interface)

I’m looking for your specific pain points:

  1. Is there a tedious analysis, routing, tracking, or categorisation task in your job, hobby, or daily workflow that existing mainstream apps handle terribly (or charge way too much for)?

  2. What’s a problem where you wish you had a smart tool to calculate, optimise, or predict outcomes for you?

Open to ideas across logistics, finance, audio/media, environmental tech, gaming analytics, or niche hobby workflows. Would love to hear what bugs you !

  • Lasherz@lemmy.world
    link
    fedilink
    arrow-up
    3
    ·
    6 days ago

    Audio custom box fabrication is currently reliant on either clunky web tools for computing resonant frequencies, port recommendations, bracing requirements, etc. There is a very nice app called bass box pro, however it’s so old that it requires compatibility mode and has a comically outdated GUI. Even within this app it does not account for complex shapes, cubic feet of the room or car open air, ideal locations for loading walls relative to the listener, direction of the cone, effective of distance from cone to port, Massloaded Vinyl thickness to bounce what % of the energy or what the floor on a downfiring sub is made of and if you need a pad for it, effects of constrained layer dampening on the inside of the box, different infill material than polyfill, thickness of the wood by wood type, final prediction weight, tips and hints on how to slide each stat up and down, so like if your maximum space allotment would give you a resonance of 37hz, what would you change to raise or lower it within different constraints like space, port dimensions, rigidity, power, etc.

    It’s heavily math-based, but the current solutions are behind the times and while it’s easy to find explanations for each question, having a self enclosed app would be great. Also FYI bassbox pro isn’t cheap, so it’s got actual value and will not really expire if the solution is global. Bassbox allows you to enter databases of new speakers to keep it relevant. Allowing car cu ft, speakers, building material specs, or something crazy like an stl or step file to accurately describe a shape or a box would be fire

  • Lasherz@lemmy.world
    link
    fedilink
    arrow-up
    2
    ·
    6 days ago

    A program which contains or accepts a database of yeast characteristics and computes likely chemical reaction favorabilities given temperature, alcohol content over time, fermentables availability, volume, and time. An example might be, “This recipe process will produce undesirable phenols” when you use more sugar than a 6% strain can use, which will stress it out and release bad flavors. Each yeast has tolerances. Also given temperature you could make predictions about desirable phenol metabolite flavors and aromas such as banana, cherry, raisins, etc. You could also do a worst case likely scenario on ambient temperature needs for a fermenter of a given size to resist heating as a result of yeast activity, so if you’re targeting 60 degrees “banana” zone, you would need to have it in a fridge at 50 degrees or whatever during the first few days of fermentation to keep it from going higher into other flavors.

    This is complicated and may require research papers. The internet can often be wrong about brewing tips and explanations for results, so I’ve found scientific literature or some Belgian brewing books like, “how to brew like a monk” to be good solid information that isn’t based on nisconceptions.

  • PriorityMotif@lemmy.world
    link
    fedilink
    arrow-up
    2
    ·
    6 days ago

    I want to find the best price for things before I buy them but I might spend more if it is closer to my house. I might also look for an alternative to that item. Search engines don’t show all the retail stores in search. I created a custom bash script which would insert my search term into the URL of each store on my area and automatically open a new tab for each website with the search. I still have to manually go through and decide but it beats going to each individual website and searching manually. Basically I would like pcpartpicker but for everything. The one I haven’t figured out yet is car parts. They have made it difficult to easily search their websites other than rockauto.

  • HubertManne@piefed.social
    link
    fedilink
    English
    arrow-up
    1
    ·
    6 days ago

    Most things I have encountered is about companies allowing for business practices to allow for software with expanded capability. for example I should be able when logged into or even at a branch of my bank dictate if money can flow into or out of an account and limit it. If that were so most people if they are smart would limit their savings to only allow for with drawls that are transfers going to checking.