Obsvra

Your data, your way, always inside your premises.

[Obsvra @Obsvra]

It's here! 1.0.0 is live NOW https://t.co/ynsiMvxNka

[Obsvra @Obsvra]

Quite true! The models shown in these videos are simplified for easy demonstration, but that is absolutely something you could extend the model to simulate for you.

That's the general idea: SpiceGrinder isn't about providing off-the shelf models that claim to be "right" -- it provides the toolkit for designing (and then running) models that match specific situations.

A more realistic (and useful) model would consist of multiple stores, multiple stages in the supply chain, and contingency options for when parts of that chain have issues.

[Obsvra @Obsvra]

Episode #5: Managing Inventory

You stock an item that sells an average of 8 units a day. It takes 2 days to resupply from the warehouse, so you re-order once your stock hits 16, right?

If you do, you wind up with more frequent out-of-stock events that you'd like.

This episode looks at simulating a store and a warehouse, then adds topics like a second store using the same warehouse or whether giving each store its own warehouse works better.

Video by @claudeai Opus
Music by @suno
Audi enhancement by @adobe_podcast

https://t.co/VPqBy5XqSH

[Obsvra @Obsvra]

@fadihares95 @grok If the completed app was worth $2000 to you, then it's worth $2000 to you. Had the developer put in 100-hour weeks to deliver the same result, would you offer to pay more?

The value is in the thing, not the effort.

[Obsvra @Obsvra]

@adnaannasir_ Lesson: do not play Calvinball with something that can think a million times faster than you.

[Obsvra @Obsvra]

I think this is the difference between programmers and engineers.

Programmers see coding as a craft. They dedicate themselves to it, and "elite coding" skills is their measure of success.

Engineers want to MAKE things. We program because that's how the work gets done. But in the end, it's the product that matters to us, not the code itself.

Agentic programming is toxic to the former group and liberating to the latter. The amount of progress I've made on SpiceGrinder in less than 4 months is staggering, and orders of magnitude more than I managed in years of it being a hobby.

And I'm having FUN doing it. Seeing my vision come to life and then evolving further beyond what I originally hoped is a thrill, and while I'm hoping for commercial success with it, just being able to BUILD it has been a joy.

[Obsvra @Obsvra]

Hey @Apple, SpiceGrinder Free was submitted for Mac App Store review almost six days ago (just a few hours shy of that), and it is still "waiting for review", which I take to mean the review hasn't even STARTED yet.

Is this what I'm paying a developer fee for?

[Obsvra @Obsvra]

@romxdev Preach it, brother.

Whenever I see someone pitching their idea based on the latest development fad, my first question to them is always "why?".

Sometimes the answer is good. Usually, it isn't.

[Obsvra @Obsvra]

@eoinmauricedaly Does nobody back up anything?

[Obsvra @Obsvra]

@GhostOfSocrates

[Obsvra @Obsvra]

@TheWuhanClan @elonmusk @KatieMiller Ponzi schemes actually work for the first few people in.

[Obsvra @Obsvra]

@kr0der A computer itself is nothing more than a giant collection of switches.

Small things assembled well and in numbers can do amazing things.

[Obsvra @Obsvra]

Episode #4: A Founder's Runway

You have $120k cash and an average burn rate of $8k per month, so you'll still have $24k remaining after a year, right?

Except 30% of years result in running out of money.

This is an extended version of an episode in our "The Simple Answer Is Wrong" series that also shows the effects of stress months in the results.

Video by @claudeai Opus
Music by @suno
Audio enhancement by @adobe_podcast

https://t.co/5Efpj2AtzR

[Obsvra @Obsvra]

@suno Have you considered adding a "generate sheet music" option?

[Obsvra @Obsvra]

@DorianDevelops At API rates? I could do it pretty easily on Opus medium.

[Obsvra @Obsvra]

@instantricecook @drefrajo Personal mouse pointer for presentations was the first thing I thought of.

[Obsvra @Obsvra]

I'd start by not writing a prompt at all.

Create a document. List the goal first and foremost, and then any actual constraints.

Break the problem down into meaningful and manageable sub-components, with a brief rationale on why each component was identified and how they should fit together.

THEN my first prompt would be to review the plan and look errors, gaps, weaknesses and to make corrective suggestions.

[Obsvra @Obsvra]

Episode #3: Simulation - a walk-in clinic and and emergency room

As we saw with the grocery store example, the average workload breaks down during peaks or rushes. We see that same behavior expressed in different ways in these two examples, which are more complex.

The walk-in clinic shows multiple stations with flow mapped between them. The emergency room adds shared resources being contended for, along with prioritization based on the specific customer.

In both cases, the simulation model and experiment tool help identify the best system improvement you can make.

Video by @claudeai Opus
Music by @suno
Voice editing with @adobe_podcast Enhance

https://t.co/FZLgyngsSt

[Obsvra @Obsvra]

@ChShersh Failing fast is a side benefit to rapid development. If the cost of trying ideas gets low enough, it's simpler to prove out which ones work than it is to decide whether or not to even try.

[Obsvra @Obsvra]

@UziCryptoo 20% down payments helped a lot.

[Obsvra @Obsvra]

@larsencc I'm loving it. Making things was always my goal, coding was a means of getting there. Now I can just direct what I want to have done instead of having to get down into the weeds of it myself. It just takes discipline to keep work units small enough to direct properly.