DEV Community: Santiago Botto

The day I became a Programmer: tales of a DevOps Engineer with Impostor Syndrome

Santiago Botto — Wed, 03 Dec 2025 17:49:33 +0000

At some point along the way, without even realizing it, I essentially became a programmer.

Once upon a time...

I was working with servers, installing and reinstalling OSs, dealing with package managers that were always causing issues when having to work with dependencies that were either too old or too new, manually setting up backups at the FS level, fighting with some servers providers because they were too slow to deploy new hardware, and a long etcetera...

I was a Linux sysadmin, then Windows sysadmin as well, and then a Cloud Architect because everyone wanted to migrate to THE CLOUD and have their own EC2 instance because AWS was the current buzzword.

Some time after that, I became a DevOps engineer, started managing infra as code, but it wasn't really programming, it was simply writing HCL and sometimes some bash scripts.

And then, at some point, everything started being either a YAML file for Kubernetes deployments, a JS script for APIs or proxies, or a Lambda function written in Python for some backend stuff.

This proposed a paradigm shift...

Why would you need a server when everything is serverless? Why would you even need a backup when you simply have a file that you host on GitHub and deploy to either a k8s cluster or a platform like Cloudflare Workers?

Has something failed? Has someone manually updated something and broke everything? Just redeploy! Is the platform having issues? Just redeploy somewhere else! You can now move on with your day like nothing happened.

Sometimes I even feel like there's no longer a need for "the infra guy"; there's always a platform to do whatever it is that the dev needs, so what's the point? But then, the issues appear and you need someone who knows DNS, for example, and you're glad you have someone that was were there at the beginning of times (aka "the hosting provider era").

Back then, we lacked several safety nets that we have today, and had no self-service platforms like Cloudflare to package everything nicely into a single thing. We were essentially full stack devs but for the Internet as a whole: DNS, Email, HTTP, DB, backups, and so forth; we handled it all with nothing but bash scripts and a prayer. The closest thing to Cloudflare I had when I started was a DNS cluster built with cPanel in DNS-only mode, and docker was just a person employed in a port to load and unload ships.

The times, they are a-changing...

I still don't fully feel like a programmer, though...

I usually feel stuck in a limbo of not knowing exactly where I fit in, or what's my place in this "new" Internet.

I'm writing Python when I work on KeeperHub, and Go for my RPC proxy (aetherlay) and some other tools I'm building, but still feels a bit off-brand for me, as if this wasn't where I belong and I'm just posing (which I think would be considered as "Impostor Syndrome").

It seems that everything has turned into code, slowly but surely, and I guess I'm here simply due to the fact that I don't fight the current... If the IT trends go in one direction and that direction makes sense, I flow with it. Be water, my friend, as Bruce Lee used to say.

How Time Machine saved my life*

Santiago Botto — Mon, 01 Dec 2025 16:00:06 +0000

*not really, but it's a nice title

I have been postponing setting up Time Machine for a while, even though I had already bought a 2TB external SSD specifically for this... For some reason, I thought that wasn't gonna be a good solution and was waiting to get a NAS set up at home.

Time went by and I just completely forgot about the NAS, and my SSD was just sitting idle at the back of a drawer. A few days ago, I was given a file with several security recommendations and general best practices for my daily life as an IT nerd, it was an initiative from one of the projects on which I'm involved since they wanted to make sure everyone working in it was as careful as possible with their digital assets.

I was already doing most of what they were suggesting, but two things really stand out for me:

Using tools by Objective-See, which are Open Source versions of some of my go-to tools for securing my PCs.
Setting up Time Machine backups.

As soon as I saw "Time Machine" in there, I immediately felt drawn to that forgotten 2TB SSD and got straight to setting it up. I configured a secure password for it (it's a SanDisk Extreme Pro, so it comes with hardware-level encryption), and then followed the very simple steps that my OS was showing me for setting up daily backups to this drive.

From now on, I simply have to connect the SSD once a day and let Time Machine do its thing.

Fast forward to 3 days after the initial set up and I found myself missing several files from my local aetherlay repo. Turns out that, while doing some cleanup, I decided to delete my local repo and clone it from GitHub to avoid pushing some changes that ended up being a bit trash (I messed my local git history). In doing so, I deleted several files that I had excluded from being tracked, files where I write stuff like pending features, guidelines to some AI agents that I sometimes use and such. Yes, for said feature list I should've been using a project-management tool, like Jira or even Notion, but I can't do that for the AI agent guidelines.

Time Machine to the rescue! In a couple of minutes, I had already recovered all my files thanks to an extremely convenient UI. No hassle of any kind at any time.

If you have some extra storage lying around, I urge you to put it to good use and set up Time Machine with it; your future self will thank you 🙏

Automated invoicing for remote workers

Santiago Botto — Sat, 16 Aug 2025 23:41:17 +0000

Ever since 2020, remote workers have become more common, and we usually work as contractors, which involves invoicing our "customers" on a monthly basis. Every month, we send the exact same invoices to the exact same people, over and over...

A few days ago, my boss said something like "please remember to send your invoices, and if you haven't automated it yet, why haven't you?" and I was like "yeah, why haven't I?".

So I did: https://github.com/sanbotto/auto-invoice

I've set up an automated way to send my invoices on a monthly basis while paying $0. No charges for hosting nor email API, thanks to these services:

Cloudflare Workers for the automated script.
MailPace for the email API.

Anyone who knows me knows that I love Cloudflare Workers and it's mainly because of this, because they have a more-than-generous free tier for you to set up anything you can think of, without having to worry about high bills.

When it came to picking an email provider, I did a bit of research trying to find privacy-focused email providers that also had a free tier. I didn't wanna just hook up to any email provider that was gonna start selling my data to 3rd parties or anything like that as soon as I signed up... I noticed several people talking about MailPace, so I reached out to them to ask for access to their free tier and they were kind enough to give it to me (they might do that for everyone who asks, but still).

I hope this simple tool can help out someone else stay on top of their invoicing.

Catch you on the flippity flip 🏀

Running Claude Code inside your dev containers

Santiago Botto — Thu, 14 Aug 2025 22:58:25 +0000

Following up on this other post, I recently started running Claude Code inside my dev containers as well. Turns out I sometimes need a bit more from my dev environment than what my sandbox can provide.

As a non-root user inside an ephemeral container, running heavy tests that require a whole bunch of dependencies is not too convenient... Sure, I could give that user more permissions, but reinstalling dependencies every single time I spin up a container can get annoying quite quickly.

I decided to handle each distinct use differently:
• Quick, disposable tests? Sandbox with ephemeral containers.
• Longer sessions working on complex projects? Dev containers, a persistent setup.

A common long-lived session for me is working on aetherlay, where I prefer using a dev container because it also lets me manage service dependencies like Redis, which I have to run in a "companion" container.

But enough talk, let's get to the code...

Here's a minimal devcontainer.json to mount your local .claude folder into your dev container so that Claude can keep and reuse its memory:

{
  "name": "Dev Container",
  "service": "dev",
  "workspaceFolder": "/workspace",
  "remoteUser": "root",
  "mounts": [
    "source=~/.claude,target=/root/.claude,type=bind,consistency=cached"
  ]
}

For the corresponding Dockerfile, you can do something like this:

# Use whatever base image you want
FROM alpine:latest

# Install Node.js and npm (required by Claude Code)
RUN apk update && apk add --no-cache nodejs npm

# Install Claude Code
RUN npm install -g @anthropic-ai/claude-code

# Match the workspaceFolder set in devcontainer.json
WORKDIR /workspace

Place these two files in a .devcontainer folder at your project root in order to have them automatically picked up by VS Code or compatible forks like Cursor.

For a more complete example, check out the config I have for aetherlay.

There's a lot of ways to do this, this one is simply meant as a starting point.

Happy coding 🖖

Replacing dshackle

Santiago Botto — Tue, 12 Aug 2025 14:26:47 +0000

I wasn’t happy with how dshackle handles RPC endpoint health checks, so I built my own proxy/load balancer:

👉 https://github.com/project-aethermesh/aetherlay 🚀

No more wasted credits 💸 on chains where you're still using Alchemy, Infura, or any other third-party provider. With aetherlay, you can:

Configure multiple chains and providers
Set your own health check intervals
Avoid code-level workarounds when an endpoint fails

The cherry on top of the cake: Want to save even more on your monthly bill? You can disable continuous checks entirely, aetherlay will only run on-demand health checks when needed. Full details in the docs 🧐

Lots of improvements are already in the pipeline, so star the repo ⭐ and watch for updates 👀

Sanboxed Claude Code

Santiago Botto — Mon, 11 Aug 2025 22:56:07 +0000

In case anyone cares about it, you can run Claude inside a container if you want to make sure to only give it the context of the codebase on which you're currently working, not your whole OS.

This isn't news, nothing that I'll post here is gonna be blowing anyone's mind, I simply want to share my take on specific topics.

Create your own sandbox image. Dockerfile shown here for reference, but can be whatever suits you as long as it has npm for installing Claude:

    FROM node:24-alpine

# Create a non-root user to run the code
RUN adduser -D claude

# Install Node.js, npm, and bash (as root)
RUN apk update && \
    apk add bash nodejs npm && \
    rm -rf /var/cache/apk/*

# Install claude-code globally (as root)
RUN npm install -g @anthropic-ai/claude-code

# Set the working directory
WORKDIR /workspace

# Give the non-root user ownership of the workspace directory
RUN chown -R claude:claude /workspace

# Switch to the non-root user
USER claude

# Set bash as the default shell
ENV SHELL=/bin/bash

# Add custom aliases for the claude user
RUN echo "alias ll='ls -lah'" > ~/.profile && \
    echo "source ~/.profile" > ~/.bashrc

# Default command to run when the container starts
CMD ["/bin/bash"]

Run an ephemeral container that has access to your current workdir + shared global settings for Claude:
```
docker run -it --rm -v $(pwd):/workspace -v ~/.claude:/home/claude/.claude claude-sandbox
```
Here, /workspace matches what you defined as WORKDIR in the Dockerfile.

(Optional) Create an alias by adding this to your .bashrc or .zshrc:

alias claude-sandbox='docker run -it --rm -v $(pwd):/workspace -v ~/.claude:/home/claude/.claude claude-sandbox'

That's it. You're not gonna have to worry about AI going rogue and messing up your whole system just because you got too deep into "vibe coding" and told Claude to do anything it wanted as long as your weekend project ended up working fine.

PS: Claude Code requires either getting a paid subscription through Claude's dedicated dashboard, or purchasing credits via Anthropic's dashboard. I didn't wanna spend 20 bucks for the monthly subscription so I decided to add money into my Anthropic's account as I tested and ended up spending $15 in 30 minutes. Don't be like me, just pay $20 upfront for the monthly subscription if you wanna try it out. Worst-case scenario, you don't like it but can still get a lot more value out of those 20 dollars.