First results from my work on the YouBot - Localization based on lidar and wheel encoders. My algo is still over-confident, and I need to improve it. But just getting all this shit to work together in ROS was so hard...
For now it's in Python, but when I am sure of its correctness I re-write it in C++ and ROS2-compatible

https://youtube.com/watch/...

We got a list of new candidates for PhD positions in the lab. There are a lot of people from the Middle East, and it made me wonder if there might be an issue with me being an Israeli.

So if there are engineers here from the Middle East, and Muslim countries in general, would you see an issue with working with an Israeli?

I guess schools in DE are very safe, because this kind of behavior is normally punched out of you around 4th grade.
Just close the fucking window and shut the fuck up instead of snitching in a chat where the boss can see. And why the fuck was that guy in the lab on Sunday?
BTW, there is no chance anyone could pull our 75+kg robots out of that window. Not to mention they are like 20 years old and their only worth is as scraps.

So I made an ancient robot work with ROS2, and revived the motion capture system so I can collect (and sync) Lidar scans, odometry and GT pose. But the worse is ahead of me...

I hit a milestone in my PhD journey today. I wrote my advisor an "I'm so done with this shit" e-mail for the first time!

I want to rant about FaceBook ads. If you go through my likes and pages I follow you'd be certain I'm a teenage boy - anime, robotics/CV labs, motocross clubs, sport car brands, skateparks, punk bands....
What ads do I get? Underwear that absorbs your period, cosmetics and wedding dresses. It's like they totally discard my personal preferences and show me ads that would appeal to the generic 30 year old female.
That's lame.
Fuck, I get ads for "all body type" clothes with plus size models. All that deep learning shit, and you can't see from my images I have a 60cm waist? At least show me the skinny girls brands....

Does anyone have experience with aligning 2D Lidar scanners?
I have a front and back facing ones, each with 270 degrees scanning range, so they have overlap.
Is there a golden standard for offline Lidar alignment, like Charuco board for camera alignment?

My modus operandi 😂
Taking my own decisions is so scary.
In the wise words of Tsunami Bomb, Independence doesn't start when you leave home!

So the battery of our YouBot died, and since it's deprecated we had to ask the EE department to manufacture one for us. When a member of my group asked them for the custom battery last year, it took them 5+ weeks to get it.
Today I went there, and threatened the engineer to cry daily outside his room until it's ready. He already ordered the parts and said I will have it assembled by Tuesday.
I am totally in love with the terrifying effect a crying female has over male engineers.

I just saw a LinkedIn post where some guy suggested to fix gender bias in Google Translate by a few lines of JS code replacing every "he" or "she" by "he/she". Damn, give that man a Turing award, he solved the problem of biased models!!!
I love that his solution is completely contextless and will allow men to finally have periods and give births. Also, that the idea of bias embedded in models in general is completely lost on him.

My ex-colleague from Sony wrote to me to consult about SLAM system for his side project. Bro...I'm reading papers for like 2 weeks and in no position to advise anyone 😂

Since I am vain as fuck but also lazy, I check my appearance in the webcam instead of walking to the bathroom and using the the mirror. As I am also technologically challenged, I'm still not sure how to open the webcam on Ubuntu, I wrote a 10 lines c++ program (OpenCV) to do it for me. Yeah...

So here I am going over the list of papers I need to read before I start working on my PhD project, and some of the papers are seriously MEH. I take a look at one of the video submissions, and of course to prove their fairly weak point, they rendered women jumping with their cleavage out and with skirts short enough to see their underwear.
I just started and I already feel like I want to puke all over these intellectuals, like that kid in the Korn video.

The hype of "we will have autonomous cars by 2025" reminds me the 2017 hype of "we will have autonomous cars by 2021". I guess if you try this PR trick enough times, eventually you will get it right.

During an annual review process, my friend, who is a female dev (doesn't work with clients), got feedback that she needs to smile more.
I'd like to hear from the male devs here, how many of you were asked to smile more?

I am preparing for my exam in a software engineering course, where we discuss best practices for programming, testing methodologies and project management.
One of the topics is CMMI, and basically it states that organizations with mature development processes can produce exceptional software even if the teams involved are average.
Do you really think an efficient process can make up for lack of brilliance/ingenuity among the devs?

Making subtle and insignificant changes in every section of the article draft is the academic equivalent for urine marking.

Apparently some trust fund kid from my uni is now a CEO - I guess it's nice when your daddy buys you a startup for Christmas.
Makes me wonder if I wasted my good years working hard on my skills instead of getting myself a "daddy".

Update:

IBM Doors. Fucking nightmare.

From today's PhD interview:
Prof: What can your teammates do that will really annoy you?
Me: Have loud, long, private phone calls in the open space. I don't want to listen to people divorcing.
Prof: And what would you do if your colleague did that?
Me: I'd walk to him....and tell him..."I understand why she is divorcing you!"

So according to Pfizer, ~39,000 people received a second dose of vaccine/placebo. So about 20,000 got the placebo, and out of that group only 85 got infected in the span of 3.5 months, during the second wave. That's 0.425%.
So they either:
* Proved Covid is not infectious
* Manipulated the shit out of their data
How that fuck is this considered valid analysis?!
If I tried to publish a paper based on such unreliable, statistically meaningless data, I would be rejected without a blink.

Good things come out of software engineering.

My best friend supported her husband for several years, financially and emotionally. Her job allowed him to focus on his training, and also allowed her to move with him to the US. Yesterday he got his UFC contract.

I am so incredibly proud of them. As engineers, we rarely get the spotlight, but we have the ability to help our loved ones go very far.

I need to stop sending my third personality to interviews. I'm just creating unrealistic expectations.

A lot of people think that the Johnny Depp vs the Sun is just gossip column garbage, but it exposes a harsh truth - a man's life can be ruined just because a woman accused him with 0 evidence, and while she proves to be an unreliable witness on multiple accounts.
Everything that was done to promote equality is being destroyed by cases like that, and I am fairly certain that many male employers will prefer not to employ women, because it's not worth the risk. And for the first time in my life, I can't even blame them.
Working in this industry is hard enough as it is for women, but this sort of shit makes it impossible. Thanks MeToo, for completely losing your mind, and screwing up women and men at the same time!

I found that being a (good) software engineer and a girl is such an effective way to piss off a lot of people. I could never pass on an opportunity like that.

For those of you that want to work in Unity3D, there's an exhausting interviewing process ahead of you. I totally feel like

I did a pretty nasty derivation of third order approximation for first derivative by hand. I got all the nasty Newton coefficients right, and was very precise until THE VERY LAST LINE, where I butchered simple math operations.
I wonder if there is a name for my particular brand of stupidity.

My interviewer described the team as very experienced, senior algorithm people, so to make sure we don't waste each other's time, I said I don't feel like a senior. He went silent for a bit. And then he said "It's so refreshing to see some modesty in our field."
😂

I've been doing some Codility exercises, and I can't see how it reflects on my skills as an engineer. It's just math riddles that can be solved with those fancy calculators.
It completely misses the aspects of managing resources, good design, understanding other people's code, hardware optimizations...