AMA: I'm an Eng Director with a long career in tech

Had registered on this site hoping to find someone to join the HR team (recruitment co-ordinator) but found this forum and I realise I can be of more help to you all here than anything.

My background: 25 years in tech, working in everything from CMS, content, sports, music, web performance, web security, observability. Can program in TCL, Perl, Java, C, C++, Go, Rust, JavaScript. Currently managing 19 people directly and 8 more indirectly (I manage managers as well as programmers). I also hire, but at the moment am looking to hire in the Americas as my team are global / remote and I need to balance the on-call rotas... but applicable to you is that I interview around 8 times per week and know what I'm looking for in candidates.

If anyone has questions they want to ask someone in the industry, go for it.

If a mod wishes to verify that I actually am in the industry just DM me.

What rising tech do you suggest learning?

Depends on the area you wish to focus. But universally data structures and algorithms are applicable to all fields, data volumes have become big (for various definitions of big) and I would say that most systems and software engineers should be aware of distributed systems (failure modes, consensus) and things like probablistic algorithms and ways to break problems down into smaller problems (map reduce).

For languages, Java is still the bread winner in large companies, but Go and Rust are what I have seen absolutely dominating startups for the last 5 years. Python is the lowest common denominator for data engineering.

With experienced hire, do you care about their degree?

Nope. Degrees are irrelevant for landing a first job (though they can lubricate the ease of gaining internships to make the first few roles more accessible).

At a certain level (after a few years experience)... all things are equal, and then it's up to the individual to carry on their learning throughout their career. What makes a senior or principal engineer isn't taught at Uni... the only thing that really carries on is your ability to research a problem, find a solution, and apply it. Experience really is "proven ability to learn and apply what you've learned".

Do you work at amazon? (your previous post seems to implicitly mention the leadership principles)

I do not. I was offered a Principal Engineer role at AWS almost 9 years ago, but I declined and I do not work for a FAANG though I have been offered roles by all but Netflix. I presently work for a startup.

What are the advantages/disadvantages of junior devs leaving big tech for a startup with higher TC?


Startups below 250 employees give you a very broad exposure to a lot of different problems and work. They accept any working solution rather than perfect, thus teaching you the value of your work is when you ship it. The pay is now equal to larger companies, so it's less a risk than it used to be and comes with a stock option lottery ticket. But there is risk, and you will be asked to do a lot of work that just has to be done and may not be fun... in startups the staff clean the toilets. That statement isn't perfectly true, but work wise there's no-one else to do it so everyone has to muck in... it's both the advantage and disadvantage in one. You can own your career potential and direction more.

In a large company you will frequently be a very small cog in a very large system, and have far less control over the work you do and how it's done. You'll learn one part of the system and will remain there a long time. The money is fine, it's stable, and if you were lucky to land in finance it's a career for life (if you can handle the burnout and disposability in the first 5 years). This is the safe option, but the one that one provide the best options for growth. If you land a FAANG then so long as you can always progress (there's no hiding) then your compounded benefits (RSUs) can be many multiples of the salary of a startup (but still potentially less than the risky lottery ticket).

Swings and roundabouts... depends what is important for you. If it's a mortgage and stable life... go large corp and big tech. If it's growth and learning, interesting work, and to pay off a mortgage sooner but start later, go startup.

thanks for the reply,

interesting that you rejected an L7 role, are you at a startup that has IPO'd ? do you have any opinion on apprenticeships for software engineering?

note: you'd want to quote people when you're replying so it sends them a notification for it

That sounds like a rut, and yes some companies only provide repetitive work that doesn't allow growth. What I'd recommend here is to look at other opportunities. Note that you can be hired for 2 reasons, 1) What you do already (your performance), and 2) What you could do (your potential). So in this case as you are in a rut you definitely do not want to be hired based on your performance, so you should focus on being hired for your potential. You already have experience enough to get an interview, so really the difference is in the interview phase, this is the filter that lines up the work that will be more challenging (as well as the best opportunity to set their expectation of your seniority, i.e. how much they will offer as a salary).

So... in interviews, how do you show your potential?

Pretty much 2 things work here and ideally you find a way to display both:

1) A bias-to-action... that you see work that needs doing, and you just get on and do it. That you ask forgiveness not permission, and that others in the team can rely on you to leave everything better than you found it. Look for ways to answer questions that hint that you see what needs doing and you actively do it... perhaps it's "to fix this I needed to go learn X and once I learned that I applied it and then showed others how to do it too", perhaps it's "when I write this frontend code I also make sure to go update the documentation and let the sales and support people know it now behaves differently". Appear to be (and actually be) someone the team and manager can rely upon.

2) Curiousity / growth... that you aren't satisfied with just the code as it is, that you always want to learn how something works and how to make it better. This is also a drive, or a passion for the work, that everything is a learning exercise. Caveat: Don't be someone "always learning", be someone who applies what they learn (that bias-to-action above).

So yes... apply for different roles that give you more opportunity, but given that you're in a rut the roles you should focus on are the ones that are a challenge to you and out of your comfort zone, and specifically you need to be hired on your potential so make sure to sell yourself in a cover letter, in the interviews, on the basis of your ownership / bias-to-action, and your curiosity / growth... those things spell potential and will help you to be given more interesting work.

- what should i learn to make good income in this year? my background is javascript and laravel
- how can i check start up company have a good future, or a company with good environment work? i wanna get rid my procastination.
- and how to learn something fast, i try to do a test interview in hackerrank, and i realize im not that good filling each test, mostly need moretime and sometimes i need to peek many documentation. I think thats make me a bad employee bcause slow minded and notgood enough for moving fast company

On apprenticeships: Check that the company is serious about it being a long-term job interview that has real potential to lead to a role. Most large company internships are low quality. Most small company internships are cheap labour for short-term projects. You need to find the ones (minority) where the work is real, challenging, a learning opportunity, and leads to a role... the best way to do this is cold-call the right manager at the right company (find them on LinkedIn, or ping someone on Github working on a project and ask for their manager's contact details) and have a cover letter that aligns your potential and interest to those teams. Managers at <2k people startups usually have carte-blanche capability to take on interns at pay that is in the £30-40k range for up to 6 months without burning their headcount allocation, and the growth means that it will lead to a role.

So I'm in favour of them... but most are poorly run or abusive (cheap labour).

Hello, you are mixing up two completely different terms. An apprenticeship is not an internship.

I just wanted to make that clear as the points you've made above do not correspond in the slightest, to what apprenticeships are. In terms of them being 'cheap labour', poorly run, abusive or only for short-term projects.

Hi @stem_leader
Thank you for starting this post.

I have been working as a healthcare professional in a clinical setting for the past 5 years but am looking at a career change into the tech industry.
I have done a few intro courses in python and web dev covering html css JavaScript basics and really enjoyed them.
I’m now planning to try to learn as much coding as I can over the next 6 months to start building a portfolio before taking part in a boot camp to accelerate the learning process, and then hopefully apply for junior software engineering roles. Do you think this sounds like a feasible plan?
I was wondering whether you were able to give some advice on whether that seems like a good way to break into the industry considering I don’t have a background in tech at all?
I was also considering doing a MSc computer science conversion course instead of a coding boot camp, as an employer which one would be more favourable?
Which coding language would you say I should focus on learning first? I’ve read that it’s better to really focus on learning one and then the learning process for other languages will be easier?
Thank you so much in advance for your help.

I'm so glad that you're enjoying programming, the only way to make a career in tech be successful is through enjoying the work.

So... boot camp. Employers do not value these highly as there is a problem... too many of these boot camps exploit those signing up by taking their money and promising to teach a whole swathe of things in 3 months, and provide a tightly structured course that do have you create something. So why is this a problem? Because it's like learning Spanish from a phrase book where your replies can only be given if the thing you are asked is "just so"... exactly as per the phrase book. You haven't learned Spanish, you've learned enough to give the illusion of speaking Spanish, but it only works if the person you're in conversation stays within tight parameters and they never do. For an employer boot camps have only introduced you to something, but not in a way that you can easily transfer to whatever the problem is that the employer has... you haven't learned to program, you've only learned to mimic a programmer. What makes this worse is that the boot camps are very profitable from selling this as a "change career and land a high-paying job in tech" and so you're many thousands out of pocket with no real increased chance of a job on the other side. I'll say that there are a few notable exceptions to boot camps, learning groups, etc... but the exceptions are so few and rare that most employers will just default to assuming that a boot camp certification means very little - that you can create a blog or other simple thing using only a very specific set of tools... which doesn't translate to the needs of the employer.

MSc on the other hand... Employers value this highly no matter which institution issued it... in fact some Unis that you may not expect have a good rep in some specialist areas, but any CompSci MSc is going to stand you in good stead. The reason here is that you will have been forced to learn, forced to stick with it, forced to stoke your curiosity and go the extra mile... all of those skills are transferrable to a business environment. The added benefit that you'll be aware of the fundamentals of computing, how networks work, and lots of other subtle things (state machines, etc)... these are the foundation on which everything we do is built, and a boot camp and self learning do not fill those in as much as an MSc does (even though at the time you probably will feel like the MSc will gloss over things).

Your general plan: learn, develop evidence that you are able to learn and apply that learning (through a portfolio), seek some qualification (MSc or boot camp) and then to seek work... this is sound. For de-risking the change in careers this is good.

I would rule out the boot camp, instead opting for either the MSc (preferred) or internships if they are available (just apply, the criteria is very fluid on internships - also, apply even when they aren't advertised as internships can be invented by employers).

As for languages... you may not yet know your interest area and what you want to work on. But as a very rough gist (extremely rough, basically so abstract as to be unusable):
* Ruby (and Rails) opens the marketing industry
* Python opens machine learning, scientific research, big data, and publishing
* Java opens big businesses
* .Net opens the financial sector and lots of small and medium business
* Go opens most startups
* Rust opens some niche and advanced startups
* JavaScript you need for frontend work in any industry, but also helps with some small companies and small business sites
* PHP opens the charity sector (Drupal and Wordpress are everywhere)
* C/C++ opens heavy industry and manufacturing

Those are very very vague generalisations... but can help steer and guide.

I personally think that what engineers should learn early in the career is a strongly-typed language which exposes some pointer and memory management (but not so much that you'd shoot yourself in the foot) as that is a simple path into other strongly typed languages whereas it's hard to go from duck-typing and interpreted languages to a strongly-typed compiled language. So I would recommend Go which you can try online: https://tour.golang.org/ It's a server based language, but teaches you all the basics of programming and actually does a good job with harder concepts like concurrency (how to run multiple jobs at the same time).

This is information overload, if you have specific questions my DMs are open.

Thank you so much for taking the time to respond in so much detail. So informative and insightful!
I’ll definitely check out everything you’ve mentioned.

Really interesting to hear that a MSc is so much more favourable than a boot camp aswell. There seems to be so many bootcamp success stories online especially from Makers and Le Wagon graduates.
I was considering the MSc computer science conversion course at the University of Westminster as it’s the most affordable out of the MScs but was worried about the uni’s low ranking, but guess that doesn’t actually matter too much!

A CompSci MSc will help you in the future too, pretty much no matter where it's from (unless it's London Met, don't go there).

A bootcamp may open the door if you're lucky enough to pick a good one and they can place you. Makers is one of the better ones but it's no guarantee at all. But the real problem is that it isn't helping you in the long run, you'll hit a glass ceiling and won't have either the qualifications to open other opportunities or the depth of learning to support yourself in that growth.

So I’ve had a look again and it’s a MSc Software engineering one year conversion course that they offer at Uni of Westminster. The curriculum looks similar to other computer science courses. Do you think that would be fairly highly regarded?

Yep definitely makes sense with your point about hitting the ceiling with a bootcamp.
So I think it may be a better plan to learn as much as I can myself, and work on personal projects whilst doing a one year conversion course, then look for jobs after that!

I am yep! I’ve been doing a few intro courses with Code First Girls and it’s such an amazing supportive community. So many super passionate people!

I’ve just realised the MSc at Westminster is not accredited by a professional body.. maybe that’s why it’s for affordable than most other MSc conversions. But hopefully that shouldn’t matter too much considering what you said about the high demand?
Or do you think I should definitely go for a course that is accredited? As long as I’m able to get a job at the end of the course it would be worth it, because I’ll (hopefully) constantly be learning on the job and improving along the way.