This is actually a good reason for exiting the industry before one's job goes away. Steering AI to barf up the right-looking pool of vomit is not the Flow-generating experience that many people have started to program for.
During COVID, I built a quick no-code solution to help ‘all-inclusive resorts’ in Cancun and Punta Cana. It started with guests scanning a QR code, filling out a short form, and scheduling their Antigen/PCR tests. Over time, it evolved to help nurses manage operations, streamline test results, and send certificates via email.
What began as a DIY project during lockdown scaled quickly to Mexico City, the Dominican Republic, and Barcelona, processing over 30,000 tests the day after New Year’s Eve. Using Airtable, an Airtable extension called miniextensions.com, and tools like Docusign, Make, Sendgrid, and Twilio, the solution ended up supporting over 1 million tests. I handled all technical aspects, support, and training myself.
Since the solution was meant to be temporary, I prioritized speed and reliability. Airtable’s relational database capabilities were crucial, especially compared to spreadsheets, for managing multiple linked tables and automations. Airtable also offers a REST API and a pretty amazing self- actualizing database schema that can be copied into tools like GPT or Claude for added context. You can build powerful tools with its scripts, interfaces, and automations.
That said, Airtable is best for use cases with fewer than 250,000 records per month. Above 100,000 records, the web platform’s performance can suffer, though there are workarounds. It’s perfect for small to medium-sized businesses needing custom API integrations for CRMs or internal tools.
On a related note, today I built a cool Airtable automation using ChatGPT and custom scripts. By filling out a corporate email in a form, the system scrapes the web for details like company name, address, phone, fax, and more, which then populates the CRM automatically. I’m still experimenting and plan to make a YouTube video about it soon. Here’s the loom video I sent to a friend about it: https://www.loom.com/share/b9ddbefbdce5434da378667fc2079d00?...
Not only do they have a fantastic full cast reading of the book, but for each chapter they have a companion episode where they talk to a Joyce scholar about the chapter. I went through it by alternating listening to a chapter then listing to the episode about the chapter, and I finally really got Joyce and understood why he is considered great.
I would not call it challenging, but it was a fun and interesting experience.
Learning functional programming through the CS 3110 course at Cornell University
Here is the link:
https://courses.cs.cornell.edu/cs3110/2021sp/textbook/
The OCaml part is free, well structured and does not take long to complete.
We're a VC-funded startup building an automation platform for Infrastructure-as-Code, adding a Policy-as-Code layer above it, in order to make IaC usable in bigger companies, where you have to take care of state consistency, selective permissions, a usable git flow, etc.
On the backend we're using 100% Go with AWS primitives. We're looking for backend developers who like doing DevOps'y stuff sometimes (because in a way it's the spirit of our company), or have experience with the cloud native ecosystem. Ideally you'd have experience working with an IaC tool, i.e. Terraform, Pulumi, Ansible, CloudFormation, Kubernetes, or SaltStack.
Overall we have a deeply technical product, trying to build something customers love to use, and already have a lot of happy and satisfied customers. We promise interesting work, the ability to open source parts of the project which don't give us a business advantage, as well as healthy working hours. We've also got investment days on Fridays, when you can work on anything you want, as long as it could possibly benefit Spacelift in some way.
Maelstrom [1], a workbench for learning distributed systems from the creator of Jepsen, includes a simple (model-checked) implementation of Raft and an excellent tutorial on implementing it.
Raft is a simple algorithm, but as others have noted, the original paper includes many correctness details often brushed over in toy implementations. Furthermore, the fallibility of real-world hardware (handling memory/disk corruption and grey failures), the requirements of real-world systems with tight latency SLAs, and a need for things like flexible quorum/dynamic cluster membership make implementing it for production a long and daunting task. The commit history of etcd and hashicorp/raft, likely the two most battle-tested open source implementations of raft that still surface correctness bugs on the regular tell you all you need to know.
The tigerbeetle team talks in detail about the real-world aspects of distributed systems on imperfect hardware/non-abstracted system models, and why they chose viewstamp replication, which predates Paxos but looks more like Raft.
Single Table Design is the way forward here. I can highly recommend The DynamoDB Book [0] and anything (talks, blogs, etc) that Rick Houlihan has put out. In previous discussions the author shared a coupon code ("HACKERNEWS") that will take $20-$50 off the cost depending on the package you buy. It worked earlier this year for me when I bought the book. It was very helpful and I referred back to it a number of times. This github repo [1] is also a wealth of information (maintained by the same guy who wrote the book).
As an added data point I don't really like programming books but bought this since the data out there on Single Table Design was sparse or not well organized, it was worth every penny for me.
What I learned in a year on Duolingo I picked up in a month of in-person non-native tutor sessions. What I learned in a year of non-native tutor sessions I picked up in a month of conversations with natives through apps like preply/italki. I wish I would've just started with preply/italki and been thrown to the wolves. Likely would've been speaking (not fluently) the languages I've learned in months instead of years.
Timothy Snyder: The Making of Modern Ukraine. [0] gives you invaluable understanding about much more than just Ukraine.
MIT 16.885J Aircraft Systems Engineering, Fall 2005 [1] - the aircraft they focus on is the Space Shuttle. Amazingly demystifying. Some of the actual early designers talk there.
Just watch the first lecture and you won't be able to not watch the rest. It starts with making your own autograd engine in 100 lines of python, similar to PyTorch and then builds up to a GPT network. He's one of the best in the field, founder of OpenAI, then Director of AI at Tesla. Nothing like the scam tutorials that just copy-paste random code from the internet.
IMO, the clear choice for this is Math Academy. It was started by a very, very early Uber contractor who built the early versions of their dispatcher, their profiler and also the grid system that helped them scale and pretty much saved them during new year's in 2013. Since becoming wealthy from Uber's IPO, he's been pouring money into hiring experts and building the curriculum.
The first step was a non-profit program started as part of the Pasadena public school system's offerings that other schools are free to adopt. It's taken students from basic arithmetic to calculus by 9th grade and through a fully undergraduate curriculum by the time they finish high school. His son was one of those students and must be part of why he's been willing to invest so heavily for so many years.
There's now a commercial online version open to the public. The founder hasn't done any marketing of it yet, but I found out about it through a mutual friend / podcast co-host. Math Academy is very comprehensive and the most streamlined way I know of to learn, or in my case, relearn an undergraduate applied math curriculum. It's not as polished, but the content and the actual academic results make offerings like Brilliant.org look like a joke.
I requested a life-time deal last year for access myself and intend to make the most of it, likely in binges between busy periods at work.
This is not really related to what this author means as "good" at chess. But I found spending about 8 hours working through lessons and chess.com and Magnus Carlsen's app (I preferred the chess.com one personally), and maybe another 8 hours playing online was about what I needed to have the occasional "good game". Relatively speaking.
It was eye opening for me learning some basic strategies and tactics. Mostly controlling the center, a couple of openings, and techniques for checkmate.
Too much beyond that and most players are doing a lot of memorization and pattern recognition. Which I don't really want to do, but I am super happy I learned the basics because now chess is fun and I understand it.
I'd recommend taking lessons online if you want to learn chess. Skip many games.
Not the person you're asking, but I run the same thing. Here's my setup:
Plex/Jellyfin for watching content.
Sonarr(TV)/Radarr(Movies)/Lidarr(Music)/Readarr(Audio/Books) - For searching/organizing/starting downloads, monitoring for new releases and so on
nzbget/qbittorrent - for doing the actual downloads
prowlarr - to handle various indexers/search providers between *arr apps and downloaders
overseerr - to make everything foolproof for people who are not tech savvy. They can see what's downloaded, and request new stuff and they get an email once its been downloaded.
On top of that I have watchtower which does automatic updates, as everything is running as docker containers.
Honestly it seems like complicated setup, but it's really not, and once set up it runs without any issues. I've had this for a few years now, and there's been maaaybe a handful of times where I needed to fix something.
