I shouldn't be doing this on funded OurBigBook time which is until the end of May, but I was getting too nervous and decided to start a casual job search to test the waters.
In particular I want to see if I can get past the HR lady step without toning down my online profiles. If nothing works out for the next round I'll be hiding anything too spicy like:Another interesting point is to see if French companies are more likely to reply given that Ciro Santilli studied at École Polytechnique which the French worship.
- prominently seeking funding for OurBigBook on my LinkedIn profile
- CIA 2010 covert communication websites references. This will be my first job hunt since I have published that article. Wish me luck.
- gay Putin profile picture on Stack Overflow
Gay Putin, currently used in Ciro Santilli's Stack Overflow profile
. Ciro's profiles may be a bit too much for the HR ladies who reject his job applications on the spot. To be fair, perhaps not enough years of experience for certain applications and job hopping may have something to do with it too. But since they don't ever tell you anything not to get sued, we'll never know.I'm looking in particular either for:
- machine learning-adjacent jobs in companies that seem to be doing something that could further AGI, e.g. automatic code generation or robotics would be ideal
- quantum computing
- systems programming, which is what I actually have work experience with
I spent the last two weeks doing that:
- one week browsing everything of interest in London and Paris and sending applications to anything that seemed both relevant and interesting. Maintaining an application list at: Section "Job application by Ciro Santilli".
- one week on a very laborious but somewhat interesting take home exercise for Linux kernel engineer a Canonical, makers of Ubuntu.I had a week to finish 5 practical coding and packaging questions, and I tried to do everything as perfectly as possible, but I somewhat underestimated the amount of work and wait needed to do everything and didn't manage to finish question 4 and missed 5. Oops let's see how that goes.At least this had a few good outcomes for the Internet as I tried to document things as nicely as I could where they were missing from Google as usual:
- I re-tested Linux Kernel Module Cheat and made some small improvements. Things still worked from a Ubuntu 24.10 host (using Docker to Ubuntu 22.04), and I also checked that kernel 6.8 builds and GDB step debugs after adding the newly required config
CONFIG_DEBUG_INFO_DWARF_TOOLCHAIN_DEFAULT, also mentioned that at: Why are there no debug symbols in my vmlinux when using gdb with /proc/kcore? - I contributed some simple updates to github.com/martinezjavier/ldd3 getting it closer to work on Linux kernel v6.8. That repository aims to keep the venerable examples from Linux kernel module book LDD3 alive on newer kernels, and is a very good source for kernel module developers.
- How to compile a Linux kernel module?: wrote a quick Ciro-approved tutorial
- Dynamic array in Linux kernel module: I gave an educational example of a dynamic byte array (like std::string) using the kvmalloc family of allocators
- quickemu: this is a good emulator manager and I think I'll be using it for Ubuntu images when needed from now on. I wrote:
- How to run Ubuntu desktop on QEMU?: an introductory tutorial to the software as their README is not that good as is often the case. It's hard for project authors to predict what new users want or not. This is my second answer to this question, the previous one focusing on a more manual approach without third party helpers.
- How to share folder between guest/host? (Quickemu): I explained how to setup a 9p mount to share a directory between guest and host
- Error :: You must put some 'source' URIs in your sources.list: updated this answer for Ubuntu 24.04. This issue comes up when you want to do either of:which don't work by default, and my answer explains how to do it from the GUI and CLI. The CLI method is specially important for Docker images. Since Ubuntu doesn't offer a stable CLI method for this, the method breaks from time to time and we have to find the new config file to edit.
sudo apt build-dep sudo apt source - What is hardware enablement (HWE)?: I learned a bit better how Ubuntu structures its kernel releases for each Ubuntu release
Some of the main issues I had were:- compiling Linux kernel for Ubuntu is extremely slow. I was used to compiling for embedded system with Buildroot, which finishes in minutes, but for Ubuntu is hours, presumably because they enable as many drivers as possible to make a single ISO work on as many different computers as possible, which makes sense, but also makes development harder
- my QEMU setup for Ubuntu was not quite as streamlined and I relearned a few things and set up quickemu. By chance I had recently come across quickemu for testing OurBigBook on MacOS, but I had to learn a bit how to set it up reasonably too
- I re-tested Linux Kernel Module Cheat and made some small improvements. Things still worked from a Ubuntu 24.10 host (using Docker to Ubuntu 22.04), and I also checked that kernel 6.8 builds and GDB step debugs after adding the newly required config
viewdns.info Created 2025-03-11 Updated 2025-07-16
This is the most accessible DNS database online, as it does not require login or payment.
They have reasonable data. It's not fully complete as Ciro Santilli saw on CIA 2010 covert communication websites, but it is very valuable.
Tested as of 2025, they seem to have removed the pre-IP checks on web interface, and just instead use Cloudfare to check that you are human from time to time, which allows for a lot manual searching to be done! Awesome!
Previously, tou could only get about 250 queries on the web interface, then 250 queries per free account via API. They check your IP when you signup, and you can't sign in twice from the same IP. They also state that Tor addresses are blacklisted. They also normalize dots in gmail addresses, so you need more diverse email accounts. But they haven't covered the
.gmail vs .googlemail trick.Their data is also quite disjoint from the data of the 2013 DNS Census. There is some overlap, but clearly their methodology is very different. Some times they slot into one another almost perfectly.