Appliku

Server Management, designed specially for Django.

Visit Website
September 5, 2022 Reached 500 Users

Today was the point when 500 people trusted us with their deployments!

That's a massive milestone for Appliku.

Over the past 12 months, we have been working hard on the stability of the app, fixing bugs, and finally dedicating a noticeable amount of time to marketing: social media presence, sponsoring email newsletters, and SEO.

Also, we've been working on the https://speedpy.com Django SaaS template and https://findtalk.social - a tool to monitor Reddit posts.

4 Comments

  1. 3

    Well done guys 👏👏👏 By far the best server Management app out there, that deserves much more credit. Hope another milestone post coming soon ;)

  2. 1

    congratulations, which channel is the most effective?

    1. 1

      Sponsoring Newsletters and reddit were the best so far

September 9, 2021 Launched Team Plans

Up until this point, Appliku only allowed Personal Projects, where only one developer could work on their apps and manage servers.

Appliku had customers that outgrew personal projects. They needed to invite their teams to work on projects together. Sharing login credentials between multiple people is a bad idea.

In early July the work was started on Team plans and it took a bit more than 2 months to roll out Team plans.

Team plans are also based on the number of servers and the number of team members is unlimited on all plans.

Release announcement: https://appliku.com/post/appliku-teams

Comment

December 15, 2020 Defined Positioning, New Logo, New Landing Page

In March 2019, I came up with the idea that I want to build a better deployment solution specifically for Django Projects.

While I was building it, I got distracted many times by other shiny things, looked at other players in PaaS space, and at some point lost my focus.

But a couple of months ago, I took a break from building the product and started heavily using it and talking to other people.

I realized the truth that building a general product for everyone means to build a product for nobody. How many times have I heard this? Only now I realized it.

I talked to people building PHP products and Ruby.

At some point, it came to my mind that I can't serve them well anyway because I am not experienced nor interested in that niche. That means I can't build a product that will appeal to them.

And finally, I got back to my original idea that I started from: Build the best deployment solution for Django Projects.

Here it is.

Also, we have a new logo and landing page with proper messaging.

https://appliku.com/

14 Comments

  1. 2

    I really like the design and Ui of your Landing page - its so clean & slick!

    1. 1

      Thanks so much :) it’s third iteration and it took us forever to ship it :)

  2. 2

    Congrats on the refocus, Kostja. It definitely makes sense to go down that road and be able to target your customers. Look forward to hearing more from you.

    1. 1

      Thanks so much, Matt! These words and support really mean a lot!

  3. 2

    Your new website is really clean, congrats!

  4. 2

    The new logo looks clean.

    1. 2

      Thanks, I can refer person who made it!

      1. 1

        I can who made this 😅
        She is talented for sure. 🙏

  5. 2

    Can only agree on your point and decision.
    Congrats on new logo and landing page

    1. 2

      Thanks ma man.

  6. 1

    Congratulations! I can completely relate to the 'pain' of deploying Django apps.

    I liked your website, it is a neat design!

    However, at first inspection, and going through the docs, there's still some questions. For example, what do you actually do? Are you going to provision a server for me? Including updating packages, setting up server, reverse proxy, etc.? Your pricing includes the DO/EC pricing? Why is only the /media persistent? In the free version you say no custom domain, but if it's on my own server, why would you have control on that? What about setting up block storage on DO, cron tasks?

    Also, the pricing seems too high. At 9/month plus the server you are going to clash face to face with Heroku, what would be the advantage of using your service? The way I read it, it is as limiting as Heroku even if running on my own server.

    1. 1

      Hello, Aquiles!

      Thanks a lot for your questions.

      There were definitely some things that I need to update in the documentation and these holidays are the perfect time for that.

      Let's start with what Appliku actually does.

      You provide credentials for your account in AWS or DO. From the Appliku panel, you create a server of the size you want. Appliku creates a server and sets up everything needed to run the app: Nginx, certbot, docker, and stuff.

      When you created an app in appliku (picked a GitHub repo, branch, and server you want to deploy to) it will make YOUR server clone the repo, builds it with the interpreter version you specified, and uses Procfile in the root of your repo to build a docker-compose to run your app. You can select what lines from Procfile to enable.

      Also, there is an option to create any number of databases, again – on your server. You can pick from Postgresql, Postgis, Redis, RabbitMQ.

      When you add a custom domain, a virtual host for this domain will be added to NGINX pointing to the app. SSL certificate will be issued with help of Let's Encrypt.

      Now about pricing.

      A single server can be as big as you want and you can fit as many apps on it as you want. And you will still pay $9/mo.

      Since you have mentioned Heroku:
      Heroku charges for each dyno a minimum of $7/mo.

      If you have a typical Django setup where you have a web process, a celery worker, and the beat worker it is already $21/mo.
      Add a PostgreSQL instance for at least $9/mo.
      Add a Redis instance for at least $15/mo.
      CloudAMQP: $19/mo

      It is already $64/mo for a single app!

      If we go with Appliku + Digital Ocean:

      Appliku single server plan: $9/mo

      Droplet 4GB 2vCPUs: $20 /mo

      It is $29/mo (vs $64/mo on Heroku) and you have plenty of room to grow for this app and for other apps too!

      By the way, we have a feature Heroku Config Vars Sync that will mirror config vars from Heroku to Appliku app.

      Using this feature one of our customers moved all their background celery workers to Digital Ocean, got rid of fat expensive dynos, and saves around $7200/mo.

      And the whole migration process took 30-40 minutes.

      Block Storage: We don't have any use for that at the moment.

      Cronjobs: coming soon.

      In the free version you say no custom domain, but if it's on my own server, why would you have control on that?

      Appliku is a service, that helps you with boring DevOps stuff, that otherwise could be done manually.

      On the free plan, it will not do a custom domain setup for you.

      Also, right now there is no support for "/media" and any persistence in the app at all, again – need to update docs for that.

      Appliku is an opinionated tool about how apps should run and I am convinced that media data should be persisted in S3.

      I am still thinking of an option for persisting data on the server, but this feature needs a lot of work to be flexible enough. As you rightfully asked: "why only /media"? So since the demand for this feature in its current state didn't exist at that time and I wasn't planning on expanding its flexibility, for the time being, it is gone.

      Let me know if there is anything else I can add to these explanations or if you have other questions too!

      Will be happy to see you among customers or just talk about your deployment needs. Hit me on Twitter via DMs and we can talk!

      1. 2

        Thanks for the detailed answer!

        I think all what you just explained to me should be part of a blog post, a walk-through guide or something like it, since it is very valuable.

        Now I have a good idea of what you do! I'll definitely keep your product in mind. Sadly, I struggled a lot to convince people to move first to custom Django (instead of Wordpresss) and then to Heroku (I am volunteering at a climate-change knowledge exchange project). And I don't have the bandwidth for another tech-change.

        I am actually thinking if there's something I can build during this Christmas break to test your service... Let's see!

        1. 1

          Hey! just noticed the reply sorry for the delay.

          Glad I was able to explain :)

          Yeah, I will save this text, polish it a bit and make a blog post about it. Thanks for the idea :)

          Let me know if you need any help when deploying your next big thing!

August 17, 2020 Posted a story of the app, got 10 users overnight

Yesterday I have published this story of Appliku on dev.to: https://dev.to/kostjapalovic/tired-of-deployments-built-my-own-heroku-47ed

It gave me very positive feedback and +10 new users.

I am pleasantly surprised by how this awesome community welcomed my post.

I did my best to make it about the story, not any kind of promotion.

With all the details about how it works and what pain I went through – it was very well received. I love developers!

2 Comments

  1. 2

    Nice, probably makes sense to onboard them personally and help them start using the app. What do you think?

  2. 1

    Cool! How many customers total?

July 26, 2020 AWS Integration Added

During conversations with potential customers there was a question I heard multiple times:

  • Great product, but does it support AWS or only Digital Ocean?

So today I am proud to announce that Appliku supports deployment to AWS EC2 instances.

There is a lot of work to be done in updating interface and onboarding, but the functionality is there and apps being deployed on AWS as simple as on Digital Ocean!

Comment

June 21, 2020 Got featured on BetaList

A couple of months of waiting after submitting to BetaList, Appliku.com became featured on BetaList.

https://betalist.com/startups/appliku

This is the first time Appliku got featured anywhere, will post about traffic and registrations when I have the data!

14 Comments

  1. 2

    Great job! Let us know if the feature returned any conversations!

    1. 1

      BetaList visitors yesterday: 34
      IH visitors yesterday: 40
      well... not much at all. But it was Sunday.

  2. 2

    Congratulations!. Good luck

  3. 2

    Hi Kostja,
    I'm using Django right now hosted on GCP. it's costing me around 9 Euros a month. The cost comes mostly from the SQL database. I can deploy with a single command and the app instance scales up and down depending on the traffic.

    If I'm moving to your service, I'd have to pay 9 euros + 5 euros for the droplet.
    Maybe I'm missing the point, but why would you say it'd be better for me to move to your service?

    1. 1

      First of all, I am happy to learn more about your setup, that's super interesting.

      If it works for you - I'd say go for it. Evidently, at this stage, it seems to be great and no point in changing anything.

      Appliku's selling point is convenience. When you are skilled enough, you can host on GCP or EC2 and get away with pretty nice prices.

      I can deploy everything manually as well. But I don't want to do it. I don't want to write configs, run a command to deploy the app, I want to add DB or change configs from the interface. I don't even want to touch Nginx configs and SSL certs – all this can be automated and that's the idea behind Appliku.

      Recently, I needed to set up 5 apps really quickly, moving them partially from Heroku. It was done in 20 minutes, including the process of deciding which workers we needed to move to Digital Ocean.

      Was it possible to do it manually? Yes. Would it be faster? Definitely no.

      As for hosting prices – the Digital Ocean was chosen as a first cloud provider because of its popularity. There are plans for adding other providers in the future. And their pricing can be different.

      At your level of cloud provider spendings, I don't think there is any reason to move anywhere – you would spend more time.

      When you need to scale, add more apps and maintaining control over your infrastructure – then you definitely should give Appliku a try.

      1. 2

        Thanks @kpalovic. I understand your value.

        I have the following questions:

        • Is this using containers under the hood or are the apps deployed directly on top of the droplet?
        • how do you handle database migrations and (in general post install commands) ?
        • is the database hosted on the same droplet ? is there a regular backup?
        1. 2
          • Docker containers
          • Appliku uses Procfile (like heroku) and it has "release" command that is executed during the deployment. You can specify command to be executed. Or you can choose to disable it in Build Settings for each app.
          • Currently database is hosted on same droplet. I chose this way for now, since I am not ready to fight in the niche of Production Grade DBs offer by Heroku, Digital Ocean and others. This is the fight for another day. Regarding backups – not yet. Backups are automatically turned on for the droplet on D.O. side. I've got plans to have backups. DBs docker volumes will be backed up on a D.O. Space. That's the high-level plan for this functionality, but I have no date for this right now.
          1. 2

            It makes sense!

            I think if you can make a cli version of your tool, open source or one time payment, you can really become some sort of standard, then monetize subscription-based the dashboard, monitoring features, backups etc

            Deploying Django apps on Digital Ocean is a lot of constellations to align. However both are used a lot by indiehackers and go through the pain and end up with some sort of variant of the setup.

            I don't like doing it and we all take shortcuts to either save money or time. If your tool focuses on being the most simple, secure and best practice fed tool, I'd buy it instantly.

  4. 2

    Congrats. Does featuring there help you with your traffic or conversion?

    1. 1

      Thanks!
      Not a single conversion to this hour. Around 40 visits from BetaList to this moment.

      1. 2

        Thanks for sharing. It is still a pretty big achievement😀

        Good luck.

  5. 2

    I just gave it a like 👍

  6. 2

    Congrats =)

May 29, 2020 First customer with a production app saves $300/mo

Big day for Appliku.
The first customer that hosted their first app on Digital Ocean with Appliku.

Despite the fact that I was very worried, deploying their apps went super smoothly with very little assistance from my side.

The reason to move was crazy spendings with Heroku.

Their app keeps using services from Heroku, but most of the dynos moved to Digital Ocean, saving them $300/mo

16 Comments

  1. 4

    I have been following Konstantin for years and know how hard was his path on the way to build the tool and get the first customer. This man amazes and inspires me. I will interview him and you all guys will be influenced too.

    1. 1

      Thanks, man. Likewise. You inspired me in your own way. Without your input it would happen some otherway than it is now :)

  2. 2

    Hi Konstantin. Thank you for the informative post. I wanted to ask you, how did you save your customer money by migrating them to Digital Ocean ? Did you optimize their services, redesign their infrastructure or was it another way ? Looking forward to your response.

    1. 1

      Hey. Thanks for your question.
      The thing is that Heroku's pricing is just nuts.
      Look at it yourself.
      https://www.heroku.com/pricing
      What we did is we moved all their background workers(everything excluding web workers) to Digital Ocean. On digital ocean for less price they got more resources.
      That's the tldr of it.

      For dyno a single dyno "Performance M" $250 /mo with 2.5GB RAM.

      What they got on digital ocean that will take care of all workers:
      RAM :16GB 6vCPUs Storage: 320GB $80/mo
      So that's the math.

      If there is anything I need to elaborate – please tell me, i'll happly will explain in more details.

  3. 2

    This is what we call the "Aha!" moment in products. When the customer sees the first real potential of a product. Kudos! This is an awesome moment, keep going at it.

    Best of luck, Sumit

    1. 1

      Thanks so much!
      Yeah I also wouldn't know that if I wasn't on a call with them.

  4. 1

    Congrats! I'm actually in the process of making something very similar in the .NET world. Would love to chat. Please send me an email through my account here if you would have a moment to do so. Thanks!

    1. 1

      Ping me on Twitter in DMs, happy to talk!

  5. 1

    That's a huge savings imo, well in everyone's opinion i guess :)

    1. 1

      It is just a beginning, we are working to squeezing more savings, but have to do it gradually.

  6. 1

    That's awesome, congrats!

    1. 1

      Thanks so much!

  7. 1

    Congratulations my friend. I know how much work went into building Appliku, so this is really well deserved!

    1. 1

      Thanks, Yasin!
      Yeah it was a lot of work and yeah it was exciting today!

  8. 1

    Congrats :)

April 25, 2020 App health visibility

Happy to report that I consider Appliku feature-complete according to my initial roadmap from April 2019.

This release I've done today brings logs in the application Dashboard as well as Processes Stats from docker-compose logs command.

These features were the missing pieces of the fully observable application management that Appliku is aspiring to provide.

Join me in being happy with these features and try them yourself!
https://appliku.com/

Comment

April 20, 2020 Not just Django, any Python app

Incredibly important update for Appliku users.

Back in a day, I have set out to make an app to make deploying Django apps easier. This was baked in my messaging and the codebase of Appliku.

This weekend I've worked on a relatively small code change, that brings significant change for the product.

Appliku supports any python app you might want to deploy.

Under the hood, it checks for the existence of manage.py before running "collectstatic" commands and migrations.

But not only that. It is now possible to DISABLE_COLLECTSTATIC=1 to disable it even if you are using Django.

Test on a minimal Flask app and works like a charm!

April 4, 2020 One year in and I figured out the positioning

One year ago I understood what type of tool I need in my day to day work.

I needed a tool that will make me more productive in starting and maintaining Django apps.

The biggest pain point was the deployment process, especially a painful amount of initial effort that goes into the first deployment.

So the first year I've spent making deployment functionality.

This February I was somewhat satisfied with how it works and shared it with the public.

Then with all these global COVID19 related tragedy and isolation and everything I was too tired to think straight and lost my focus.

I forgot why exactly I have started https://Appliku.com .

This week I had a bit more time than usual and started talking to several people about Appliku.

I've made a new landing page and asked them for feedback.

Everyone was mostly happy with the design, thanks to https://unicornplatform.com .

But people were very confused by what exactly is Appliku and what I plan for the future, what features I plan to add.

"Will it support other languages? What about building the frontend part? What about running nodejs apps?"

"Is it hosting, reseller or something else?"

"What's the point of using Appliku if it is not a hosting?"

"What about scaling? Can it scale my app?"

"Wait, do I need to write code by myself?"

I became super sad, confused and generally got into all these thoughts.

I started thinking that I explain everything wrong.
If so, I need to understand who is my target audience. I had feature requests(more like questions, since I don't have users yet) that were conflicting with each other due to different types of audiences.

I took a day off of doing anything in front of my computer and started thinking. Something felt missing, something was completely off. But what?

By the evening, I realized: I have missed my focus.

I was comparing Appliku directly to Heroku. I thought about how to position Appliku in a way that it had benefits when compared to Heroku.

I quickly realized that those who already use Heroku are unlikely to move anywhere until they are ready to hire a DevOps department and go with EC2 or something.

Solo/Indie makers most of the time don't have money to pay for anything so they will probably do everything by themselves. Or they will because they lack skills and time to deploy apps themselves.

Maybe somebody with revenue will want to switch to Heroku, but it is risky to do and hard to convince them. If their app is making revenue, then you probably don't want to risk everything by screwing things up in this early stage.

Enterprises are out of the question here, because I know what goes into the sales process here. I am not ready to do all this.

I have spent the whole day thinking about it and finally recalled what was the initial idea: Help Python/Django developers in the way like no generic platform can possibly help.

Just like GitLab. You may have labeled it as a direct competitor to GitHub, but it is not. It is a platform that has so many useful stuff built-in, that it can't be compared to GitHub. Especially, when GitLab provides most of that stuff for free, but it is a different conversation.

So Appliku is not a deployment app. That's not the future, it is a core feature.

Right now, I am thinking of it this way:
Appliku is the all-in-one cloud platform for creating, running and managing Python/Django apps that help boost developers' productivity and run apps with cost-efficiency and flexible scaling.

Comment

About

I didn't want to overpay for Heroku. I wanted to deploy on my own servers.