A series of Medium posts based on my new book, “An Approach to Machine Learning in Cyber Defense for the DoD”.
Technical talent has long been a topic of complaint for many officers and leaders alike in the DoD. So what do we make of it? It is a talent gap that can be solved for? If so, how? How can talent be amassed to tackle current (frankly, past) and future disruptive technologies like machine learning?
Several topics tie into the talent recruitment, management and retainment challenge facing the entirety of the public sector today. The first is the issue of having non-technical officers and/or leadership managing a team of technical engineers. Then the issue of contrast- the massive differences between what we will call “islands of brilliance” and “waters of mediocrity.” A discussion around the role of contracted technical talent must be had as well (as it is often touted as the route to solve the technical talent shortage) and finally the topic of motivators will be discussed. The next post will cap this post off nicely by elaborating on what types of talents are perceived by industry experts to be necessary to form the perfect machine learning team.
“Do we react to the problem set and capabilities that we know today… or are generations to come going to have baked in capabilities that we haven’t yet thought about how to leverage, which will make this current world of cyber completely irrelevant?”
Technical talent is scarce. Everyone is feeling this and knows this- from commercial companies like Google to the government agencies of the highest intrigue such as the National Security Agency (NSA). An agency known for its technical prestige and elite workforce such as the NSA is even struggling to find thought leaders in the technical space to remain as researchers and ultimately leadership to lead cutting edge developments in the field. Most importantly, that talent is exiting the workforce long before it can become the leaders we actually need in the field right now- the leaders who can relate to engineers, and leaders who can enforce the right type of culture where engineer-minded young talent will thrive and remain.
The Issue of Non-Technical Leadership
The importance of people who create the workplace culture, embody it, enforce it and lead it will be the ultimate determining factor of whether the DoD can truly break into the technical fields that will alter the world as the internet did but in even MORE technical and abstract depth in years to come. Workplace culture is oftentimes set by those in leadership positions, which too often include those will no technical background and therefore have nothing to work with to gain the trust and respect of employees with technical achievements that do not match those in positions of leaderships’ ideas of achievements.
I keep calling these individuals “those in leadership positions” because not every individual in the DoD (for as much as the military touts leadership) is a good leader. When it comes to technical talent, recruitment and retainment, most current DoD leaders are absolutely not what is needed to attract, retain, usher engineers through a career path, then send them off to their next life goal adequately. Never mind taking care of people- it is very hard for an engineer to trust a structure of non-technical minded decisions makers who do not even bother to ask what technical terms in conversation mean lest they reveal their ignorance on the subject.
Engineers and those with technical skills are therefore sidelined. They are not understood, and yet the DoD culture has no mandate to understand and embrace new things and new realities. While policies fly ahead of DoD personnel (with the U.S. military arguably pushing the hardest on AI and ML adoption from a policy standpoint) the employees of the DoD lag behind… far behind.
How can you expect to recruit and keep technical and engineer-minded individuals if your leadership cannot articulate themselves about common technologies used today let alone recognize the importance in the change of mission and new talent requirement we face?
This is, of course not to say that every single leader in the DoD is non-technical. I think we have sufficiently gotten to a handful. But if you’re asking yourself that question now and you lead a technical team…
Islands of Brilliance Amongst Waters of Mediocrity
There is a common saying: “government civilians are impossible to fire.” The fact that this type of statement is followed by a range of chuckles and eye rolls is unforgivable in an organization charged with protecting America’s interest at home and abroad. Why, might you ask, do I start on this seemingly obscure subject?
At the heart of what creates the still waters of mediocrity is the feeling of job security which leads to complacency. Even amongst military members is there rarely a feeling of having to push oneself to keep one’s job and promote. This feeling is especially prevalent amongst government civilians, many of whom form the immediate middle management chain of command under the non-technical military leadership layer we discussed above. And those government civilians are worried and concerned about one thing and one thing only: their job security.
Despite government civilians being some of the most secure positions in (arguably) the entire universe, for some reason this motivation prevails over their decision making. So when it comes to hiring and retaining engineers- who generally run fast, break many things along the way, and leave whenever there is a hint of a stagnant work environment- there is no way these two opposing forces can exist amongst one another.
This becomes especially impossible when these mediocre waters tend to rule over the islands of brilliance, and hold middle management and decision making authority over things like resources, time, priorities, and projects. The waters of mediocrity become so still they constrain the technical mission so much that barely any innovation and building can be done.
Now we don’t have to blame EVERYTHING on just the individuals themselves that compose the waters of mediocrity- lack of retainment in the DoD can also be directly tied to factors such as the unsuitable culture the military fosters for technical people alone with the flawed administrative practices that plague the apparatus and force its employees into a coma of complacency. Those are equally as important- but ultimately those two factors are enforced by the people hired at the DoD, so to the subject of people we must always return.
As before, this is OF COURSE not to say that every single government civilian is evil. They are not. I have had the privilege of working under a very small handful who DID fight for the hardest yet most innovative path forward. Those
Contracted Technical Talent
So- one might say (usually your 0–4 or 0–5 in charge of the technical mission) let’s hire technical people as contractors and pay them boatloads of money. That should solve the problem, shouldn’t it?
In fact, it doesn’t. It is a common misconception that large tech companies like Google, Amazon and Microsoft pay their employees … boatloads of money. While this might be true in a sense, they do NOT pay their employees boatloads of money in the sense of salary. They may compensate them by other means- more paid days off, stock, awesome health benefits, free food, etc etc… But when the numbers are compared, it is in fact hilariously obvious that a senior developer at Booz might make much, much more than his or her equivalent at the big tech conglomerates.
Here is what else contracting out tech talent doesn’t fix: it does NOT fix the broken culture and environment in which these engineers must step into and build something out of nothing. The administrative flaws are still glaring, innovation is stifled, tech is misunderstood, technical people cast aside as necessary burdens- overall not a welcoming environment, and not one where very many ideas and even prototypes will ever see the light of day.
“If applied machine learning was never really part of the company’s core DNA, this tech talent will inevitably try to find a home elsewhere…”
Contracting out talent, therefore, does not fix the problem of mission mindset, which is ultimately one of the delicate pairs of hands that cradles personnel and talent management in the DoD.
Let’s also consider for a moment the sources from which technical talent come. A large defense industry contractor who has been supplying butts-in-seats to the DoD for decades? Awesome- just send that one junior IT guy to a Python bootcamp and relabel him as a junior data analyst, and there you have it- a data analyst!
The DoD and the U.S. government as a whole needs to start becoming more creative with where talent is sourced. If the answer for a pen tester is some guy who couldn’t pass a PT test in any of the last nine lives he might have had, covered with tattoos and piercings… get off of your hill of mightiness and go hire that person. Then watch him or her work their magic.
“…The Russian government outsources its own technical talent, drawing from what might be considered unconventional areas of recruitment.”
Also, money doesn’t motivate everyone, which leads us to…
What motivates technical talent to… well, stay put? What would motivate an engineer to just stop leaving and going somewhere else?
I don’t by any means claim to have all of the answers. As an engineer myself who has hopped from many jobs (almost yearly) who interviewed a bunch of engineers with DoD backgrounds, there were some glaring points pointing to what motivates us that have to be highlighted.
Don’t just offer me money. There is more to life than that.
Engineers due to some weird reason actually care much more about other things: quality of life, work/life balance, ability to pursue hobbies and nooks and crannies of interest unhindered, freedom and independence to get a problem solved however they need to.
Why does stock work?
While stock is, yes, money- what it also does is offer the employee a way out of they find a compelling reason that is more than that stock is worth. This is much better than signing a contract that legally binds you (as a federal servant, literally) for three to four years to an organization or job position. What stock also does is it gives the engineer a feeling of ownership over what it is they are actually about to join.
While stock might not be an option for public sector or DoD positions- have leaders thought about what could be close? Instead of binding legal contracts for military members, how about cash bonuses? Being a member of a guild? Finding their community? With the number of people a prospective military member has to touch in the recruitment and onboarding process, you would think SOMEONE would have a creative idea to change this.
The mission is cool… until it’s not.
Using the DoD mission as a way to try and attract technical candidates will not do it in the long run. It might also not do it in the short run either. What is to say that someone WANTS to deploy to a war zone and leave their homes? What our own DoD organizations TEND to forget is 1. being a part of the military is not like the movies and 2. Once you find out the secrets and go through the pain of gaining a security clearance, it all becomes unexciting anyways and 3. If you know all of the things but can’t DO anything with or for them because of all of the reasons we discussed above, after a while the spirit of the engineer just becomes…. well, tired. And fed up.
Lastly… the spirit of the engineer.
Technical talent is leaving the DoD just as quickly as it is rushing in the door (then potentially out again) of large tech companies. The spirit of the engineer is oftentimes missed by BOTH the DoD and the large tech company industry. The engineer wants to build useful tools for people who delight in using those tools. The engineer will continue to roam until he or she finds somewhere where their own flavour of innovation can thrive, unhindered and unfettered. If the spirit of the engineer is not welcomed, they will continue to leave, or never show up at all in the first place.
At the end of the day, if you read this whole post and think, “Rubbish! My organization is not like that! I am not like that! I have the PERFECT environment for an engineer to find a home and tackle hard problems!” then I encourage you to read the next post outlining how to build a true machine learning engineering team to solve your hard problems with disruptive technologies.