I think this is exactly it. Replit is a cheap and easy way to get an MVP off the ground ASAP. However, their audience is inherently hackathon attendees, not real businesses. Whether these can turn into real businesses (en masse to justify low churn and consistent SaaS ARR) or not is the real question.
thanks for sharing, that's an interesting perspective actually. It's easy for us "pro devs" to kind of ignore platforms like Replit as "training wheels." I look at it and think "why would I use that, I have all my own stuff set up the way I like it locally".
But us older guys (i'm not that old, 34, but still) can easily forget how valuable and exciting it is to have tools that make the publication / deploy easy. It's cool to hear what the younger, less experienced crowd gravitates towards in the modern dev tool landscape. Thanks for sharing!
Excellent point. With that being said, I think there is market potential for replit, specifically in the middle ground between 'not knowing any code' and 'full on developer using an IDE/Cursor'.
I'm not sure why you're getting downvoted. These are valid criticisms of platforms like Lovable, Bolt, Replit, etc. that cater to "fast MVP" type customers. I'm not sure how you sustain the valuations if the churn inherent to those type of "hobbyist" or "solopreneur" type customers isn't solved.
A friend used Replit to prove out a startup (it worked) and what worked for him is that Replit has a whole platform integrated with their coding assistant that include hosting and backend runtimes. So his cycle time of vibe-deploy-test was very short and very simple for someone non-technical.
No need to think about how/where to deploy, cloud platforms (AWS, GCP, Azure), etc. Just vibe and deploy.
(He did end up moving off the platform once he had enough validation)
Most experienced programmers have no experience deploying apps (or their experience is from earlier in their career). Especially engineers at big companies where there are whole teams dedicated to infra/devops.
The percentage of programmers with side projects they deploy themselves is very small. I’d guess less than 10% have a side project deployed somewhere. (And these days
> Most experienced programmers have no experience deploying apps (or their experience is from earlier in their career)
Most experienced programmers in my circles have evening/weekend projects. We are notorious for hoarding unused domains for the "brilliant side project" that gets a burst of commits right after domain-renewal time
Yeah. I'd say about 1/4 of my time on my new app has been spent on deployment-related stuff, rather than the app itself. And I'm not inexperienced with servers and cloud. It's a pretty big deal to integrate that stuff.
I have a habit now of getting that out of the way first just so I don't have to think about it. Get a basic functioning prototype and then figure out my infra and deployment as early as possible.
Depending on my projects, I tend to keep it pretty simple.
For personal projects, usually Firebase (+ occasional Cloud Run mixed in) which makes it relatively easy.
For professional projects, it's pretty easy now on AWS with their (unfortunately named) Copilot CLI [0] (highly, highly recommended).
But mostly, I keep my infra simple and bias towards modular monoliths [1] which ends up being the majority of my infra work (container packaging and deployment of the initial runtime infra).
Both make it pretty dead simple to deploy. AWS Copilot being the "more powerful" of the two, but still dead simple to use compared to CDK, Cloud Formation, or writing Terraform or Pulumi scripts.
I've really shortened the loop on deploying my side projects with Claude Code. I run it with `--dangerously-skip-permissions` on a prompt I've written and it adapts it for the project in hand with a "safe" set of defaults, and I've got a basic verification script to ensure it's not unsafe (e.g. can't access postgres from the web, firewall blocking all non-required ports). The rest - which can vary from project to project, like creating VMs, configuring rules, whether it's a rust project or a docker compose file - Claude knows how to handle pretty well. Super super simple now.
> (He did end up moving off the platform once he had enough validation)
I'm really curious what this looks like in practice? Like can you just download the whole codebase, throw it against a Supabase Postgres DB, and you're off running? What about any backing services or microservices? Is it tied to any thing like lambdas etc.
I should be clear here that "moving off the platform" involved a re-write for various reasons. First and foremost was that the LLM generated code was in a bad state due to the fact that he started in late 2024 when coding agents weren't really quite there yet and he had accumulated a LOT of tech debt very quickly. But Replit allowed him to validate the business viability first with some absolutely trash tier code (hacked 3x; one time where the hacker event sent an email to all customers).
Not sure why this is controversial. I know it’s an issue with Cursor as they have to limit availability of models based on region. OpenAI specifically blocks India and Pakistan for example, among a long list of other countries.
Why would anyone region-block a country which gives them a ton of users? OpenAI actually has India-specific plans alongside their regular ones, and I use Claude Code every day with zero problems.
2. > Because the doctor will rigorously apply artificial and unreliable diagnostic categories backed up by invalid and arbitrary screens and queries to make a diagnosis. So after this completely subjective and near useless evaluation is completed, your doctor should be able to exercise prudent clinical judgment to decide if Ritalin could be of benefit.
What else can you do for psychiatric conditions? We don't have a magic ADHD-o-meter but know that it statistically impacts lifespan, health, etc. Even for more objective measures like blood glucose, BP, BMI, clinical interventions are based on discrete thresholds that don't exist in nature.
TLP was doxxed in a way that threatened their real life psychiatry practice, briefly blogged on Tumblr under a different psuedonym, and has since had little online presence other than rare tweets and randomly dropping a self-published book on Amazon (_Sadly, Porn_ by 'Edward Teach').
Everyone has an LLM tool a couple clicks away if they want it, so I don't think we need this kind of contribution. And this summary is too much of a summary to be useful anyways.
There is such a thing as state-dependent memory or context-dependent memory: recall is better when the environmental context (e.g. location, lighting, smells) matches the context of learning.
If you study while on Adderall, which alters your neurochemical state (increasing dopamine and norepinephrine activity), you may recall that material more effectively when you are in the same neurochemical state, that is, also on Adderall.
Similarly, if someone learns something while sober, they will generally recall it better when sober again, rather than under the influence of a drug.
It is the phenomenon where memory recall is improved when the internal physiological or psychological state matches that during learning.
The comment above reduces the interesting article to a lacking one sentence summary. It is indicative of someone who is both too lazy to read something for themselves and for some reason thinks it is a good idea to admit this publicly.
Not a college dude, but i used to work on shits (including night shifts) and adjusting to and from a five-nights (23:30-07:30) shift isn’t that pleasant either.
- i mean yes u cannot make money out of teenagers but damn replit's Vibe coding tool is fucking good. Better than Lovable or Bolt any day.
just to give u a perspective from a 20year old kid from a 3rd world county