aecs99

Hi Peter, thanks for the AMA.

Are there any benefits to converting from a green card to citizenship, in the context of employment? Access to some government jobs is the only thing I can think of.

Can green cards be extended every time they are about to expire until someone retires, or is there a limit?

Amazing idea and execution. If you don’t mind revealing, can you describe the ML components in the background? Model, data trained on, etc.? Thanks and good luck!

Please keep in mind that the tips/tools/techniques that people list are highly subjective and may not be applicable everywhere. I realized this from my own experience (worked at two startups in the past, and now at BigCo):

* Glassdoor: Sort reviews by date (people are relatively happy when they join, negative reviews tend to be more current). Take every review (good/bad) with a grain of salt. Reviews with higher "helpful" counts could be more reliable because there are so many companies where the HR writes glowing reviews

* Crunchbase: To know more about the funding situation, list of execs, etc.

* levels.fyi: Salary info. Also check h1bsalaryinfo (website). May/may not be applicable to you, but will give you an idea of what your peers potentially make.

* LinkedIn: Search the name of the company, click on the "People" tab, and look at the list of people who match your background, and then dig out their backgrounds. How long have they been working at the company, what are their previous jobs, how well do you think their background support their roles (even before you talk to them). At my previous company, there were a bunch of people who did sub-par job of being software engineers at their previous companies, and took director/lead/vp roles purely based on their years or experience rather than technical expertise and screwed up the mission big time. This tends to happen at medium/big companies more frequently than at small startups.

* On-site interviews: I know it is going to be extremely hard to gauge a team/company while they are gauging you, but this can be done. Prepare your questions beforehand, and try to see how close or vague their responses are.

* Blind (app) or teamblind (on web): Search for the company and look for comments/questions etc. There may be some discussions which may not be applicable, but it doesn't hurt to search.

* GitHub/Medium: Some companies/startups have their own pages, or at least their team members do (although not current). It'd be good to check those as well.

* Cold emailing: Previous and current employees. Not many will respond, but if even one or two do, I'd highly value their input

* Coffee: Doesn't hurt to try (most people are willing to do this, but don't do this just because they don't have time, or feel awkward about meeting new people)

Most of these will be applicable for smaller companies/startups, but at big companies, it'll be tough knowing all the people you will work with. Ultimately, it's going to boil down to few things:

(a) your instincts: whether you feel you'll fit right in or not, whether the company is good for you or not, whether you believe in the core mission or not. I tend to believe this more.

(b) your team: a good team/team member can make your life at work the best experience or the worst experience. And unfortunately, there is no way to predict this unless you give it a chance.

Good luck!

I did. Once I joined the companies that did not openly answer my questions during interviews (about finances, strike price, etc.), I ended up realizing all the negatives/problems of the companies from the inside.

Then on, I had only two choices: (1) ignore the problems and not worry about future, or (2) act fast and start interviewing until I have options if something goes wrong. I chose the second option, and hence a lot of interviewing.

So true! I've interviewed with several startups (about 50-60) in the past, over the course of 5 years. I had offers from most of them, while some of them rejected me after the interviews (for whatever reasons they had). Your comment is so close to reality (based on my interactions). Some laugh, some genuinely have no clue, some act arrogant (you can either join based on whatever limited information is provided, or leave), and some say they don't share any such information.

I've worked at two startups in the past. The first startup tanked. When interviewing for the next role, I did ask these questions, but had no luck. Ended up taking an offer with 15% increase over my last role. Two years of work, and I find out that this startup too, is on its way down. Eventually ended up moving to a big company.

Some good points by others on this post. I want to focus on something more latent. Are you sure there are no other distractions in your life?

Try to find out if there is something deeper than surface level that is bothering you. Such distractions can be of any type: lack of interest in what you do (or rather, the constant desire of doing something else), homesickness, heartbreak, financial hardship, issues with family or friends, etc. If there are any such reasons, you should try addressing them first. Either solve them, or get some help in coping with factors you cannot solve.

I personally know two people who quit high paying industry jobs and moved to reputable research grant universities (within US) as Assistant Professors in their late 30s. This was about 8 years ago. Fast forward to the present, one guy is a full professor while the other is still an associate professor. The difference was that one used his expertise and experience to write and get more grants, publish more papers, and is popular for his research (both inside and outside the university). The other is known for his teaching (within the university), took it slow and didn't publish as much.

I also know another guy who got his PhD, moved to academia (research lab and all), quit and moved to gaming industry to code, worked for 5 years, and now moved back to academia (once again, research lab and all). Then there is another person who got his PhD, worked as a post-doc, worked as an Assistant Professor, quit because he didn't enjoy it, and now is working next to me, enjoying an industry position.

I think all of them are truly enjoying what they do. I guess the question for you is what would you like the most?

There are several questions that you'd have to answer for yourself:

(1) Are you in for teaching? Or are you into research, i.e., having freedom in what you work on? Keep in mind that if you join as a Assistant Prof. on a tenure-track role, you'd still have to prove yourself in the long run. This could mean working on some projects that you may/may not enjoy in the short term.

(2) If you are in for the teaching, do you care about where you teach? Community colleges or small universities are always looking for people to teach (as a full time professor, or as a part time lecturer). Do you differentiate between these as much over the love of teaching?

(3) You probably could try out guest lecturing to check if you truly enjoy teaching. Or maybe teach just for a semester, if that is any appealing.

I personally have a PhD, wanted to be in academia for a long time but jumped to industry for numerous personal reasons. This is my 5th year in industry and I love what I work on. However, I still feel that my heart is in academia. To get a reality check, I'll be guest-/co-lecturing several sessions of a course at a public university this fall. I'm curious how things will turn out. Good luck to you too!

I agree. I don't think this isn't any more revealing. Just a different way of visualizing. And there is also the drawback of a whole new interpretation when you re-map your input features to Chernoff facial features.

I worked with Chernoff faces long time back and love how this is an interesting way to visualize how discriminative your features are.

The idea is that you take features of your dataset, and use those to represent a face. Say for example, you want to classify 100 people based on different features. And let's say you've collected 15 features for each person (e.g., height, weight, shoulder width, length of first name, length of last name, type of car driven, etc.). Now try mapping each of these features to Chernoff faces. You'd map it in the following manner: height->area of face, weight->shape of face, shoulder width->length of nose, length of first name->location of mouth, length of last name->curve of smile, type of car driven->width of mouth, etc.

Once you've mapped in that fashion and visualize the faces, you can observe how discriminative your features are. How do you interpret this? If your Chernoff faces tend to show a lot of variation in expression (e.g., smiling vs. sad), you say the length of last name is more discriminative. On the other hand, if the faces all appear to have same area, your first feature (i.e., height) is not very discriminative.

Other features used for Chernoff faces could be: location, separation, angle, shape, and width of eyes; location, and width of pupil; location, angle, and width of eyebrow, etc.

One drawback (as listed in the Wikipedia page) is that we humans perceive the importance of these faces by the way in which variables are mapped to the Chernoff facial features. If the feature mapping is not carefully chosen, your largest varying feature may be ignored because we appreciated the change in expression more than the change in eyebrow length.

I lean towards what Velodyne is saying in this situation. I have been working with LiDAR systems for over 4 years of which the last 1.5 years have been towards building autonomous driving vehicles. When I saw the videos, I was truly baffled by how a LiDAR can miss that. I worked with different types of LiDARs (from different manufacturers) and there is a very high chance that the LiDAR point cloud contains all the information corresponding to the person and the bicycle to make a decision.

What we need to keep in mind is that sensing an object is different from deciding whether or not to take an action (e.g., hitting brakes, raising alarms, swerving, etc.).

Most LiDAR/RADAR/Camera manufacturers only provide input data. It's like saying "hey, I see this". It's up to the perception software to decide whether or not to make a decision.

In most cars, relatively simpler decisions are made by the car's perception software (e.g., adaptive cruise control, lane change warning, automatic braking, etc.).

Self-driving companies override such systems, and rewire the car such that it is their perception software that makes the decision. So the onus is completely on the self-driving company's software. In this case, it is the perception software developed by Uber to be critiqued - not Velodyne, not Volvo, not the camera manufacturer.

It looks like the engineers at Velodyne feel confident that they should (and would have) sensed the person, and hence their statement. I wouldn't doubt them much as they have been in the LiDAR game since DARPA days when self driving was considered experimental.

From a different angle, Velodyne may not have much to loose by throwing Uber under the bus - especially when compared to how much their reputation is at stake. This is because Velodyne has several big customers (e.g., Waymo, and almost every other self-driving, mapping company that is serious about getting big).

NTSB should and will get access to the point clouds. Uber has a choice of releasing the point clouds to the public - but I highly doubt they will.

You summed up all my speculations in one sentence

Great question. At least in our algorithms we do this - to adjust the driving speed based on the conditions (e.g., visibility or perception capabilities).

At the end of the day, you can drive only as fast as your perception capabilities. A good example of that is how fast humans can perceive when influenced by drugs/alcohol/medications vs. when uninfluenced.

What is baffling is the fact that the car was driving at 38 mph in a 35 mph zone. This should not happen regardless of how well/poor your sensing/perception capabilities are.