Intel Officer Question

There is no such question as a stupid question here. I call this forum "stupid questions" because then you can ask whatever you want and if someone says "That is a stupid question" then you know you put it in the right place.

Intel Officer Question

Postby jriot504 » Tue Jan 09, 2018 5:54 pm

My package is complete and will be submitted to both the SWO and Intel board - 05Mar and 06Mar. In the event I am chosen for both, I put my chances at 42%, I am looking for more information regarding Intel. SWO offers plenty of information that I have read, videos I have watched and SWOs I have talked to. Intel offers little in comparison.

The main theme for Intel I have read is becoming proficient at Powerpoint. As my signature states, I am a data scientist, so I understand the analysis which goes into making the Powerpoints - the work people don't see when they are being presented to. My team has presented to universities, national conferences, the Federal Reserve, Brookings Institute etc... That aspect of Intel is not new for me - aside from the research being more life and death related.

My questions:
1) How is intel analysis completed? Qualitative? Reading, connecting pieces, knowledge gained from experience?
2) Are quantitative skills useful? Econometric analysis, spatial analysis, natural language processing...?
3) Are technical skills sought after? For example - over the last six months I have written machine learning algorithms to analyze all job postings and resumes in the United States, labor market data which is half a terabyte of text based, JSON formatted data. Is the field open to providing new methodology in seeking reliable intelligence? Would being able to write algorithms from scratch to analyze intelligence data be useful?

I don't want to come off as trying to change or do the work differently as it is a field I know little about. I am aware the Navy will train me to conduct intelligence work the Navy way. However, I want to ensure the decision, if there is a decision, as I said 42%, is the correct decision. I also want to offer the best skill-set I can to the Navy. Lastly, where can I have best career given my interests, skills and long-term goals.
  • 1

jriot504
Registered Member
 
Posts: 6
Joined: Tue Jan 09, 2018 5:53 pm
Reputation: 1

Re: Intel Officer Question

Postby Sum1 » Thu Jan 11, 2018 1:07 am

Personally, If you're a data scientist writing machine learning algorithms I'd think you'd be a better fit for Cryptologic Warfare Officer (CWO). All the Intel officers I've ever met really do spend most of their time honing briefing skills. It's a core capability. #1 is "all of the above". #2 is "when able". #3 is "go CWO instead of Intel".
  • 0

Sum1
Experienced Member
 
Posts: 961
Joined: Wed Jun 11, 2008 1:43 am
Reputation: 15

Re: Intel Officer Question

Postby yoshi » Thu Jan 11, 2018 3:54 pm

jriot -
congratulations on submitting your package and good luck. i'll take a crack at answering your questions:
1) what sum1 said - all of the above.
2) you know the answer to this question, not sure why you asked it. Of course quantitative skills are useful; qualitative skills are also.
3) again, you know the answer to this question (when aren't technical skills useful?). technical skill is the current crack addiction used to treat functionality abuse in the IWC.

i will offer a different view than my friend sum1 in that i think you may prefer the Intel community than the cryptologic community (if your goal is to apply your skills to generate quality analysis). the CW community today performs less analysis than it used to (ironically, partially as a result of new tools/systems) and that analysis focuses on a narrow set of sources (relative to Intel). as intel is broader in scope, there is more complexity, more opportunity. generally, the CW community has analytic tools and access which already outpace human ability necessary to employ them. This is what happens when you make tools, but don't provide any training. Yet, we focus on building/inventing more - we have an entire command dedicated to it. The primary factor limiting our utility, however, is related to our inability to enable human beings (standard organizational dysfunction - elevated levels of authority, inadequate training, administrative distractions, poorly run organizations, poor leadership). IWC has a lot of brainpower - it's difficult to find another organization with as much. Perhaps the commonly held perception of deficits in technical skills and tools is a symptom, rather than the problem itself.

Certainly, there is room for AI, but in order for that to be successful, we must have and maintain a complete and total understanding of the target set in order for AI generated analysis to be any more fruitful than what we have now. we have to be able to identify all of the qualities (not just quantify or engineer) of that set and intimately understand the nature of its actions as they occur. although sad, this ability within the CW community has largely degraded in the Fleet over the past 20 years (really since the end of the Cold War). i believe the intel community is slightly ahead of the CW community in this regard. as the intel community works across all intelligence disciplines (not just SIGINT) and maintains an all all source approach, more variables are available to you, thus yielding opportunity to much more finely tune your algorithms, generating higher definition for the resulting analytics. while it is true the intel community is much more powerpoint focused, there are tons of opportunities for anyone with your skill set to introduce or build a new, effective methodology.
  • 0

yoshi
Experienced Member
 
Posts: 427
Joined: Thu Jul 10, 2008 5:10 am
Reputation: 19

Re: Intel Officer Question

Postby Sum1 » Fri Jan 12, 2018 8:48 pm

yoshi wrote:jriot -
congratulations on submitting your package and good luck. i'll take a crack at answering your questions:
1) what sum1 said - all of the above.
2) you know the answer to this question, not sure why you asked it. Of course quantitative skills are useful; qualitative skills are also.
3) again, you know the answer to this question (when aren't technical skills useful?). technical skill is the current crack addiction used to treat functionality abuse in the IWC.

i will offer a different view than my friend sum1 in that i think you may prefer the Intel community than the cryptologic community (if your goal is to apply your skills to generate quality analysis). the CW community today performs less analysis than it used to (ironically, partially as a result of new tools/systems) and that analysis focuses on a narrow set of sources (relative to Intel). as intel is broader in scope, there is more complexity, more opportunity. generally, the CW community has analytic tools and access which already outpace human ability necessary to employ them. This is what happens when you make tools, but don't provide any training. Yet, we focus on building/inventing more - we have an entire command dedicated to it. The primary factor limiting our utility, however, is related to our inability to enable human beings (standard organizational dysfunction - elevated levels of authority, inadequate training, administrative distractions, poorly run organizations, poor leadership). IWC has a lot of brainpower - it's difficult to find another organization with as much. Perhaps the commonly held perception of deficits in technical skills and tools is a symptom, rather than the problem itself.

Certainly, there is room for AI, but in order for that to be successful, we must have and maintain a complete and total understanding of the target set in order for AI generated analysis to be any more fruitful than what we have now. we have to be able to identify all of the qualities (not just quantify or engineer) of that set and intimately understand the nature of its actions as they occur. although sad, this ability within the CW community has largely degraded in the Fleet over the past 20 years (really since the end of the Cold War). i believe the intel community is slightly ahead of the CW community in this regard. as the intel community works across all intelligence disciplines (not just SIGINT) and maintains an all all source approach, more variables are available to you, thus yielding opportunity to much more finely tune your algorithms, generating higher definition for the resulting analytics. while it is true the intel community is much more powerpoint focused, there are tons of opportunities for anyone with your skill set to introduce or build a new, effective methodology.


Where is any of that actually happening in the Intel community? I'd surmise that it is not happening, nor would there be an abundance of time/support for such an endeavor as the Intel Officer is coming up the ranks. Where in a career would AI algorithm development fit in? Early JO years you're generally providing direct intel support at the tactical level, later JO/early senior officer you're on staffs that likely won't have the technology to support AI, and after that you're the N2 or deputy N2 where your life is largely controlled by your boss and what their needs are.

I think there's more chance of working on such things with some of our joint tours with three letter agencies, or at the other units who do capability funding and development. I think we need to carefully differentiate between technology development and technology utilization... and I don't think intel officers are doing a whole lot of tech development.
  • 0

Sum1
Experienced Member
 
Posts: 961
Joined: Wed Jun 11, 2008 1:43 am
Reputation: 15

Re: Intel Officer Question

Postby yoshi » Tue Jan 16, 2018 3:51 pm

are you asking where all source intel fusion is occurring? if so, i'd offer CNFK (where, by the way, they have an SMRD for their own cryppies without any cryppie officers), MOCs, in many of the lovely vacation spots with which we've become familiar (AFG, IRQ), and every day onboard ARGs and CSGs (just not in the rooms with which you may be familiar). To say it isn't happening is to profer an uninformed assertion. As for employment of a JO and lack of technology to support - I think you've been around long enough to realize if you want to get something done, you have to do it on the side. Holding them up as inherent barriers is a false argument. We both know we don't have good enough tech and career development now to produce any better result than those we are "enjoying" now. problems are resolved when someone decides that they are going to do something different, something more. when officers view their opportunities as being confined to the role assigned by their billet/boss, you have the system we have. If our jobs were good enough to provide decent technology and proper employment, we wouldn't have the problems we do. I've seen JOs automate running entire exercises from excel - nobody gave them technology or training on how to do so. they did it because they saw an opportunity to apply their skills. it's much more about creativity and innovative desire rather than about what job hangs around your neck. in the case of the author of this thread, i think the intel community presently holds a broader range of possibilities for someone looking to provide the widest ranging impact to the intelligence discipline. they're better represented in the fleet and COCOMs, outside of the national intel centers.

As for tech development- i've felt for a long time the cryppie community has had more than ample technological prowess. the personnel in the stable are plenty bright and capable (evidence the JOs i referenced above). the problem isn't AI or a lack of technical background. the problem is with the system - it stagnates individual development and prevents critical solutions, as most are content to allow the limits of their billet and their boss to determine what they able to get done (the two reasons you espouse in your response). figuring out how to get the technical solutions in place is more important and beneficial than any technical solution. this is what we should be focusing on.
  • 0

yoshi
Experienced Member
 
Posts: 427
Joined: Thu Jul 10, 2008 5:10 am
Reputation: 19

Re: Intel Officer Question

Postby Sum1 » Wed Jan 17, 2018 5:53 am

I'm aware that Intel officers perform all source analysis :)

(well, some do... others just give excuses why they don't have the expertise required to do the kind of all source analysis the N3/J3 needs)

I meant the role of AI and someone who can write and optimize AI / Machine Learning algorithms. I don't know of any intel commands where that happens, nor one that would have the technology required for dabbling. That's what I was trying to (poorly) express. I think if the original poster wants to keep doing things like that as part of their job and not a hobby the CWO community really is their best bet.
  • 0

Sum1
Experienced Member
 
Posts: 961
Joined: Wed Jun 11, 2008 1:43 am
Reputation: 15

Re: Intel Officer Question

Postby CNO Guy » Thu Jan 18, 2018 7:36 pm

Rather than get what INTEL Officers do or don't do from CW Officers, I would encourage you to contact the INTEL Officer Community Manager who is responsible for new accessions into the community that person is:

CAPT Sheryl Richardson
Intelligence Officer Community Manager
sheryl.s.richardson@navy.mil
901-874-3426

If you want to talk about CW officers this is a good venue or you can contact me at:
CDR L. Seth Lawrence
CW and CWE Officer Community Manager
lemuel.lawrence@navy.mil
(901) 874-3123

Best of luck in your pursuit of getting more data and your ultimate selection.
  • 1

User avatar
CNO Guy
Experienced Member
 
Posts: 71
Joined: Mon Feb 02, 2009 8:00 pm
Reputation: 21

Re: Intel Officer Question

Postby jriot504 » Thu Feb 08, 2018 2:56 am

Thank you all for the replies. For some reason the system did not notify me until recently that there had been any responses to my question.

Since I posted this I had NAVAIR reach out for interview, I let them know about my OCS package.

Based on the responses and my interview with NAVAIR, there is one area that really sticks out. The government does not allow you to fiddle with computer systems and install software as needed. I use Linux, run code on Linux based servers and can install different languages, libraries/packages as need to test ideas. Being limited by what the government allows or requires 10 steps to get a new a package installed would impact being able to implement any AI/ML.

For example, if the department you are assigned only has programming X installed for us, but programming X doesn't have ndarray capability, then that will significantly slow down you progress of testing. You would be required to write you own ndarray data structure, which is not easy and might not have the functionality for what you need to do. This is but one example.

However, I appreciate everything that was provided!

I also have an offer right now to work as a data scientist for a cyber security company start-up in the NW. Great salary, stock in the company, work from home and be able to complete my PhD in mathematics at the state university near my house.

Tough decision I will have to make if I am selected next month.
  • 0

jriot504
Registered Member
 
Posts: 6
Joined: Tue Jan 09, 2018 5:53 pm
Reputation: 1

Re: Intel Officer Question

Postby yoshi » Thu Feb 08, 2018 1:24 pm

Best of luck in whichever endeavor you choose!
  • 0

yoshi
Experienced Member
 
Posts: 427
Joined: Thu Jul 10, 2008 5:10 am
Reputation: 19


Return to Stupid Questions about Information Warfare

Who is online

Users browsing this forum: No registered users and 0 guests