Overcoming Bureaucracy Burnout with Hannah Hunt

Hannah Hunt [00:00:00]:
Ultimately, warfighters are going to lose trust in the program office if they are not seeing updates and results based on the things that they need, and instead just the PMO comes back after a year and hasn't made any changes that they requested, but built more functionality that's not useful for them. That cycle is terrible and you end up having software and other like, hardware capabilities that are delivered that actually don't support the mission needs of those warfighters.

Ryan Connell [00:00:47]:
Hey, everyone, this is Ryan Connell with the chief digital and Artificial Intelligence office, here today with Hannah Hunt. Hannah, welcome.

Hannah Hunt [00:00:54]:
Glad to be here.

Ryan Connell [00:00:55]:
Let's kick off and just give yourself an intro. I know you've been in and out of DoD and other experiences, so can you just give a brief overview of your experience?

Hannah Hunt [00:01:05]:
Yeah, absolutely. Right now, I am a distinguished technical fellow with Metrostar Systems. I sit within our defense business unit and I support our customers across the defense portfolio. I've been at the company now for about ten months, but before that I was in front of government, which is where I had a blast working at Kessel Run, one of the software factories in the air force as the chief of staff and acquisitions program manager, and then moved over to the army software factory at a futures command where I was one of the inaugural employees and was the chief of product over there. So after spending some time in government, I needed a break from the bureaucracy and went over to the private sector. So it's good to still, still be government adjacent, I would say.

Ryan Connell [00:01:50]:
I appreciate that. Let's go in chronological. So I'm curious, I think you were part of some of the software factory startup several years ago. At this point, we'd love to just kind of understand how that operated. My understanding is it's kind of like a butts and chairs, lack of better term, but we're paying for developers, not for product. Can you talk a little bit about that? I just want to understand that arrangement.

Hannah Hunt [00:02:10]:
Absolutely. What I think was really unique about Kessel run from an acquisition strategy perspective. Really big shout out to Tori Kuff and Matt Nelson, who spearheaded that from the beginning, was, yes, it was less about delivery of some specific product, but there were requirements to deliver in a way that were agile with the government being ultimately serving as the system integrator. And to me, that was a really fascinating concept because I had worked in peos before where I oversaw basically a vendor to deliver something without me having any say or buy in to make sure that it was the correct solution. And so Kessel run from an acquisitions perspective, said, we're going to do these modular contracts mostly to small businesses and focusing on delivering capacity to build app teams that would ultimately have government oversight and government product managers, and deliver value to the user base, which is the primarily the air operations center. Airmen so I thought it was a really amazing concept of embracing both what exists currently in the far, but also at the same time leveraging at the time the middle tier acquisition authorities and now the software acquisition pathway to be able to deliver value quickly to the user base without being tied into the lengthy DoD 5000 process that really limits innovation and the ability to deliver.

Ryan Connell [00:03:37]:
You mentioned user base a couple of times. I'm curious what that relationship was like. Curious you talking to users daily, weekly, monthly? What was that iteration? How were you getting feedback? I think this is an area that we can improve upon across the enterprise.

Hannah Hunt [00:03:56]:
I think our product teams were talking to users on a daily basis, whether they were stationed overseas and Qatar or others, or bringing in users to do six month TD wise at Kessel run and really making sure that whatever we were building was going to be valuable to them. Instead of building against some big r requirement. We were looking at instead those user outcomes, those changes in behavior that were valuable for the mission set that they were working on, whether that was targeting or geoint or planning whatever kind of process that exists out of their tasking order. And instead of waiting six months to a year to do user acceptance testing that was happening continually. Some gripe that I have in general is the lengthy process that currently exists to conduct a user acceptance test or to do user testing. It often requires a memo and taskings in order to do that, versus it being something that can really be more holistic and iterative, finding some champions, some beta users that are interested in feedback and then delivering that to them in a timely manner. Because ultimately warfighters are going to lose trust in the program office if they are not seeing updates and results based on the things that they need, and instead just the PMO comes back after a year and hasn't made any changes that they requested, but built more functionality that's not useful for them. That cycle is terrible and you end up having software and other hardware capabilities that are delivered that actually don't support the mission needs of those warfighters.

Ryan Connell [00:05:32]:
Yeah, that makes a whole lot of sense. You mentioned a six month TDY, as an example, is that were you sending the user to Kessel run or the Kessel run person to be with the users?

Hannah Hunt [00:05:42]:
Both. And I think both things are palpable. But having airmen go and live in the AOCs and understand the day to day life of the targeteers or the planners is immensely powerful because you're seeing their pain points live and then can make changes and updates live versus having to go back to the dev team and then waiting to get the new ATO or whatever the security requirements are. All of that can happen because we built an iterative process that embraced rapid delivery of features into production. And that is really, I think at the end of the day you can have agile teams and do scrum like activities, but if you are still doing user testing in a waterfall manner and you're doing your cybersecurity activities in a waterfall manner, you're still not going to be able to deliver value to that user base at the speed of relevance. And that's something that Kelsey run was able to hack really early on. And ultimately I was just the people that got the contracts in place to be able to do that. And then ultimately, as chief of staff, made sure that we had the right people to do those things.

Hannah Hunt [00:06:51]:
But it was something that was so fascinating to me, and then I was able to carry on to my later roles, that embrace of quick, successive, iterative, incremental improvement.

Ryan Connell [00:07:02]:
One other thing that I want to poke on and just understand a little bit better from your perspectives. I see in my role how things like authority to operate ATO can be a challenge to get over, and we're dealing with some of the newer cutting edge technologies. You're going to see that there potentially is a lot better technology that exists commercially that just doesn't have the appropriate safeguards for DoD, and we're trying to build those in. What does it look like in a software factory as it pertains to, like an ATO? Because we build it, we have it, or what does that look like?

Hannah Hunt [00:07:32]:
Yeah, so there's definitely different ways that the software factories across the DoD have kind of implemented their security process. There has been a lot of times you hear software factory and you think greenfield development, which for the most part is true. So you're building a custom application, typically a progressive web application, but there's also refactoring of existing applications, and you need the ability, to your point, to insert new technology, new libraries, open source components that are going to provide net value to your offering. And so from a security perspective, I understand the importance of aligning with the compliance pieces because they are there for a reason. Your stake checklists, your srgs, your other eMass packages, your security plans, all of that is great. But if you're actually not implementing the real, what I consider real security components, like your security scanning container, hardening your architecture to make sure that you're encrypting everything end to end, as an example, then you're really not practicing real cybersecurity, something that the software factories have done really well, starting with Kessel run and then across both. Now the Space Force, the Navy, Marine Corps and the army, has been this concept that you accredit the platform and the process, and then as soon as those applications meet whatever security guard rules are defined by the AO or the isM, that application can go into production. So it's less about these massive stage gates and these updates to these stigs.

Hannah Hunt [00:09:07]:
But I have met, for example, if I'm a system owner and I'm trying to build an application and get that into production, I need to make sure that I have a high enough code quality. My security scans show no critical and high vulnerabilities, and among other things. And if that's true, then I should be able to go to prod automatically versus having to wait for somebody to check some paper box that says I'm good to go. And that's a really challenging thing to navigate from a bureaucracy perspective, because security is often seen as a very manual and gate based process. And it doesn't have to be, can be integrated, just like in devsecops, integrated throughout the entire lifecycle, software development lifecycle. But it requires a lot of people who are willing to make that change, and that's where organizational inertia can be a challenge and a pain point at time.

Ryan Connell [00:10:01]:
Sure. Yeah, absolutely. That's actually super insightful on the platform based approach. Let's jump into your move to army futures from a career standpoint. Just interested. You went from, I think, acquisition and chief of staff type resourcing functions to chief of product, which seems like a pretty big transition. Like, what was that like for you?

Hannah Hunt [00:10:25]:
Definitely a big transition. I got really interested when I was at Kessel, run around platforms and around product management, and I bothered a lot of people, asked a lot of questions, did a lot of my own research learning, because I was just so interested in what platforms are, but how do I build software securely and sustainably? So when army software factory was standing up with two majors at the time taking this thing by the horns, there was no idea necessarily on how we would make sure that what we were training the soldiers on how to build software, but we needed to also be able to build good product for whatever user base was out there in the army. We had to show value, and being an organization that upskills soldiers on how to build software is equally important. But then actually showing, doing and delivering was going to provide us that cachet to be able to keep going. And so I joined the team with no role, really. I was actually hired as a chief innovation officer, which is kind of a made up title from my perspective, but I really just wanted to provide value where I could. And there was this gap in how we build and deliver products to the army. And I've been really fortunate, I was super fortunate in my time at software factory to be able to build some really useful products to the army.

Hannah Hunt [00:11:48]:
Things from the. To support the reservist community on finding jobs, active duty jobs that they could apply to, to improving the maintenance and checklist process for various equipment, to improving the development of ammunition holding apparatuses in war and conflict zones, to be able to make sure you have secure ammo like these types of things are of super useful and valuable to the army. And with a group of soldiers that had just learned how to be product managers or designers or software or platform engineers, to be able to actually put something in production and gain value from that was the most rewarding thing that I was able to do at the software factory.

Ryan Connell [00:12:32]:
Do you have any background on how at some point, someone made the decision that within the army software factory, we're not going to necessarily model around what it sounds like Kessler run did, but instead bring in soldiers and teach them how to do software acquisition.

Hannah Hunt [00:12:49]:
So I think what army software factory took a lot of lessons learned from the Air Force and the build up the software factories. Some of the challenges that we saw were that the Air Force had never institutionalized a career field for airmen in this capacity. There had been some attempts with these kind of special designations, and you had an enlisted career field, but nothing. A holistic career field that the airmen who had gained these skills on how to build software could then use and build throughout their career and actually get promoted. And so a big push that we had from the beginning was, how do I institutionalize this upskilling? So it's not just a one off and they do a six month tDy, and then, you know, they go back to their normal jobs, but how do I build the entire dot mill pfemenous around a capability where soldiers learned how to build software and so that was at the forefront of our entire mission. So the software factory started with kind of this target dream that there would be a new MOS, new series that they can apply to schoolhouse structures. And at the same time, the benefit of that is then we as a software factory, could find problems throughout the army and build custom software to solve those problems. And that was my job, was to one find the problems and work with organizations to identify pain points and then ultimately oversee all the software development and product development within.

Hannah Hunt [00:14:17]:
Within the army software factory. What makes army software factory unique as compared to the other, like Air Force software factories is Army Futures Command is not an acquisition organization. It doesn't have acquisition authority. Army software factory didn't have acquisition authority. So what we tried to do is just kind of like a lab, but not an actual lab. Be able to build proofs of concepts that provided value with real users, then find partners of which we could then transition that software or sustain it for a certain amount of time and then bring it over to, hopefully into a program of record, could then utilize it. And there's been many instances where software we've built has become institutionalized. The ordinance schoolhouse, for example, uses software that, software factories built in training, which is really exciting.

Hannah Hunt [00:15:09]:
And then there's also been an opportunity to transition to different programs up to army acquisition levels. And so to me, that's a success, is that, hey, we built something useful and then it continues on regardless of whether soldiers are continuing to build it or not. But that soldier has that domain context that most people don't have. So you're able to, one, have rapid feedback loops with users, but also then be able to iterate quickly.

Ryan Connell [00:15:38]:
Yeah, it's interesting because I'm kind of tying some trends. Of course, as you'd imagine, between your kind of first two spots of your career here, where you went about it differently, but in both of your career stints there in DoD, there was significant user input to helping define problems to find requirements, be it a six month tdy to come hang with you or vice versa, or actually just having the. The soldiers who have that domain expertise, teaching them, upskilling them, to be able to start tackling software solutions, which is super interesting, right?

Hannah Hunt [00:16:08]:
And they have friends, right? So they came from a variety of moss and skill sets. And so those people, those soldiers were willing to be beta users. We found stakeholders that were interested in leveraging our capabilities and being engaged. We worked across various operational units to validate what we were building, and not just sitting in a silo at features command, but actually partnering along the way. And that was something that was unique, too. Kessel Run did that exceptionally well in the beginning in partnering with the AOC's Air combat command to find the right balance of user feedback, stakeholder feedback, aligning with requirements. And I think army software factory was able to take the lessons learned from that and be able to implement it within a future state concept of having that career field at the end of the prototype.

Ryan Connell [00:17:02]:
Yeah, I don't think I'm exaggerating by saying, like, you hit the utopia of teaching a fisherman how to fish. Right? Like, instead of, it's a. I shouldn't say it's easy, but it's a lot easier to go buy someone some sort of software solution. It sounds like you didn't do that. You taught people how to figure out how to solve problems and do potentially their own software development.

Hannah Hunt [00:17:24]:
Yeah, absolutely. And something that was really exciting me, too, is some of the soldiers for their next assignment at the software factory are going into acquisition programs. So they're taking the skill sets they've learned and now can do that for programs of record and use what they've learned, the skill sets they've gained to help make sure that whatever we procure as an army or as a DoD is going to be actually valuable and have the right oversight to make sure that it is useful. Having lived in this various program offices, I can understand the struggle of dealing with the bureaucracy and the milestones and the various compliance laws and things, but it gives me a lot of hope that, that these captains and majors primarily are going into acquisition organizations with this skill set, that the army values them to be in an acquisition role, and then they can support and making sure that what is procured from a software perspective is useful.

Ryan Connell [00:18:17]:
Yeah, that's incredible. I know you like, we've talked several times, but I didn't know this extent of the things that you've worked on. So this is really neat to me personally, and I'm excited that we have a third of this conversation. We've hit two thirds of your career. Um, so I'm excited for the third herd because you've already hit so many great things, but would love to know, you know, you finally, you said earlier, done with the bureaucracy, needed to get out. We're with Metro Star now. What's that like? What are you doing?

Hannah Hunt [00:18:43]:
I do want to say that, like doing government service, there's nothing like being in the government and being able to make a mission impact. But something that I really personally struggled with was I was emotionally investing. So much into outcomes that were certainly out of my control, things that generals and political appointees had to make decisions on. And so when things didn't go the right, you know, the way that I wanted them to go, that really took an emotional toll on me. And so I got really burnt out at the end of the day. And so I had to make a decision for myself to step away, still be able to support the mission from an industry perspective. But I couldn't keep doing to myself what I was facing. And so it was a tough decision for myself and my family to do that.

Hannah Hunt [00:19:28]:
But at the end of the day, I'm really glad that I was able to find Metrostar. I was actually going to leave a government contracting type world and go commercial because I just was fed up. And then Metrostar's culture really interested me because there seemed to be the things that I loved about government and the software factories I worked with. Having a bias for action, having a focus on autonomy and delivering excellence was something that Metro store had. I took a kind of a leap of faith and joined Metrostar about ten months ago. And I'm really enjoying my role. I get to do a lot of things. I'm in a kind of a hybrid role, which is really exciting.

Hannah Hunt [00:20:06]:
I get to support our DoD customers right now I'm actively engaging with the Marine Corps on some of their HR modernization processes. And then I also get to partner with other companies and other DoD organizations to help them try to solve the problems that they have. I spend a lot of time in government doing what I fondly called internal consulting, being that government voice to help navigate bureaucracy, find opportunities to move through things, and ended a lot of workshops and engagements with different government organizations. And now I get to do that on the outside, but also make sure that I'm maintaining my own kind of mental health from being able to diversify the type of work that I get to do versus kind of being in a one program and getting burnt out from that. I love Metrostar. I think there's endless potential as a company, but also as a people that I really enjoy. And again, I get to do whatever I want, which is really refreshing sometimes when you've been in the bureaucracy for so long.

Ryan Connell [00:21:04]:
Sure, absolutely. I remember when you took that role or soon after you had a LinkedIn post that made me chuckle. And I remember commenting on it. Everyone who leaves DoD makes the same LinkedIn post. And it was about like transparency and we're putting rfps on the street, like lack of understanding and transparency and con, like just making a lot of sense, basically. I think you had this aha. Moment of how hard we make it on contractors sometimes. Do you remember that? Do you know what I'm talking about?

Hannah Hunt [00:21:29]:
Oh, I do. I think I had a Taylor Swift if I remember correctly.

Ryan Connell [00:21:34]:
Yeah, go through that. Explain that a little bit for the listeners.

Hannah Hunt [00:21:37]:
Oh gosh. Public apology was well required. I had learned now being on this side that companies take rfis as real requirements like that things are going to come out because an RFI was put out and I probably did half a dozen rfis that I knew that there wasn't actually a requirement for in the sense of like I didn't know if I had money for it or I didn't know if I was going to be able to pull it off, that I was trying to get some market intel. And companies take that as the government's word that yes, there isn't a procurement, but they're going to put resources, business development, capture resources against those opportunities. And having received and submitted several rfis when I first joined Metrostar, I realized, oh, these aren't hard and fast like requirements at all. And I'm burning the candle at both ends to try to respond to these things when really it's just the government trying to get information, things that I have now a lot of empathy for. I also public apologize to several companies that applied to a bid that I had to put out right before Christmas, which I super sorry about that. Not intentional.

Hannah Hunt [00:22:44]:
There were things beyond my control that required that. But those types of things, we need to be seeing it from this side, need to be more mindful that the industry has families and they need to take breaks too. And it can't just be throwing stuff at the wall and hoping that you get some market research done. Things, I mean, to the CDAO's credit, trade wins is a great way to do market research and it does not require you to put on RFI. You just go to the marketplace out there and that is a great innovative way because you don't need to do RFI's to do market research. The far gives you immense flexibility in what it means to do market research. I think we're just so comfortable using rfis that there needs to be a better and more streamlined way to do it. And again, to CDAO's credit, you guys are doing that, which is fantastic.

Hannah Hunt [00:23:32]:
I think more organizations need to embrace that and get away from paper responses.

Ryan Connell [00:23:37]:
You don't need to do rfis to do market research is my underlying statement after this conversation. I love it. I love it. Curious. I didn't mean to take away from the LinkedIn thing. It just ran through my head. I was like, oh my gosh, we got to talk about this. Anything else? You talked a little bit about bureaucracy.

Ryan Connell [00:23:54]:
You talked a little bit about. We just talked about the transparency and empathy. Any other biggest lessons learned or discovery from kind of moving from DoD to industry?

Hannah Hunt [00:24:03]:
Yeah. I would also say that there doesn't need to be an adversarial relationship with industry either. I will be the first to state that when I was a program manager had a disproportionate disdain for industry initially, when really there's things that I didn't know or technologies that I wasn't aware of, and I could have learned more if I had more of an open mind. It's really refreshing to engage with government customers that want to have that information now. And I'll take the first stone versus myself, who was like, I'll just put an RFI out and I'll use that to finalize contract strategy. But really it can be something much more holistic. So that's a takeaway. And then I'll also say that something I learned in government, and what I'm bringing here too, is that I can only care as much about something as other people are willing to care about it.

Hannah Hunt [00:24:59]:
So I realized in government that when I wasn't able to get traction on a particular goal that I had and it would just burn me out. And it took me a while to realize that if that general officer that sds above me does not care, I can only go so much in trying to convince them otherwise. And I have to accept that's where we're at and I have to move forward. I'm seeing that the industry side too, around, hey, I might think this is a really good opportunity that we should go after, or I might think that this is the right delivery choice to make with the Marine Corps or others. And it may not happen. And that's okay. There has to be a level of acceptance that I did what I could advocate for, what I need do you think is best? And then there's decision that needs to be made. And this is a very Amazon thing, the disagree and commit type mindset.

Hannah Hunt [00:25:48]:
Like, okay, I disagree, but I'm going to move forward for the betterment of the company or the organization.

Ryan Connell [00:25:53]:
Absolutely. It reminds me not everything needs to be written out in policy or regulation, but it's funny to see I'm not only a year, maybe three or four years ago far part one was updated to include language that basically says the government must not hesitate to communicate with industry as early as possible in the acquisition cycle. And I remember thinking, Goshen, unfortunate they had to put that into regulation.

Hannah Hunt [00:26:17]:
Yeah, I referenced far part one an awful lot. Same with far part 39 and the modular contracting being a good mechanism for IT procurement. Oftentimes I see a disdain for that, thinking that one system integrator is going to solve all your problems. So really you need to diversify your vendor base to make sure that what is, if one vendor isn't performing, you can then realign resources to a vendor that is performing. That also requires a lot of strong oversight by the government with KPI's or quas or whatever. But yeah, I love the far and it's my passion and people don't make fun of me, but I'm like, yeah, just check out far part eight that the federal supply schedules. You can see what's possible if you want to use a GSA vehicle and it's fine. I accept that about myself.

Hannah Hunt [00:27:04]:
That's the things I like. Just like you love pricing.

Ryan Connell [00:27:07]:
Yeah, sure, exactly. We can nerd out. That's fine. Hey, anything else on the Metrostar front worth communicating or anything that you wanted to share beyond what we've talked about?

Hannah Hunt [00:27:17]:
Yeah, I'll just give a plug for the work that we're actually doing with CDAO right now. We're supporting the JTech teeny effort and our platform is the base platform for all these applications working on model development and testing evaluation for the AI assurance effort. And I think that's one of the coolest programs that we have in the company right now because we're able to deliver a baseline platform and ultimately hopefully an enterprise platform that allows us to test models as a department to make sure that they work well functionally and they work securely so that they can be deployed in whatever environment is needed. And I think that's a really awesome and unique mission of the CDAO, be able to set that standard of how we're going to do testing and evaluation for AI. And I think personally I'm super jazzed to be a part of it and with our little slice of it that can make an impact. So that's my last Metro star plug. Come check us out on our website. It's a cool company of cool people.

Hannah Hunt [00:28:18]:
I really enjoy being there.

Ryan Connell [00:28:20]:
Awesome. I'll give you an opportunity here. We're getting close to time, so anything you want to leave listeners with? Advice for people that might be potentially thinking about moving from industry to government or vice versa or even just entering into the software world.

Hannah Hunt [00:28:35]:
Yeah, I would. If you're thinking, hey, I'm burnout by the government, I need to leave. It took me a while to make that decision and I think give yourself space to find the right role and the right company that's going to meet the needs that you're looking for. Sometimes it's better to have to wait for the right company and position than it is to just choose your first offer. So I would definitely caution people there from a personal perspective, always try to grow your craft and be curious. Be interested in new technologies. The world is moving really fast. Generative AI changes weekly at this point.

Hannah Hunt [00:29:11]:
Always be interested in new technology, especially as a government person, because there is opportunity to bring this into your program that you may not realize. Ill Fitz on that. And then the last thing ill say is, for any people interested in book recommendations, I always give out a few recommendations. One is always the lean startup by Eric Reese. That's a great book to talk about how to build iterative software and be able to scale it effectively. Range by David Epstein is a great book that talks about having a breadth of experience versus just a depth in a single set of experiences. You'll be more creative. I also recommend if you're a manager, making of a manager by Julie Zhao.

Hannah Hunt [00:29:50]:
She's former chief of product at chief of design at Facebook. And it for me was super helpful when I accidentally became chief of staff and had to then manage a bunch of people to learn what it means to be a good manager. And I reference her material all the time. So for any government supervisor, any supervisor, it's a great, very tactical, practical implementation of how you can be a good manager for your people. Because at the end of the day, that's what matters. You can spend all your money and build your program, but if you're not taking care of your people, they're going to leave, they're going to find other places to go, and you're going to lose a lot of that edge that you have by bringing in smart, dedicated people who want to solve real challenges.

Ryan Connell [00:30:32]:
That's awesome. Spot on, right? You take care of the people, they take care of the mission. Um, so appreciate that. All right, question I ask everyone, as a homage, I guess, to Bonnie is, uh, as a way to sign off, is if given the opportunity, would you dip your grilled cheese into your hot chocolate?

Hannah Hunt [00:30:50]:
Yes, I think cheese and chocolate go very well together, so. Absolutely.

Ryan Connell [00:30:57]:
She's in. All right. Hey, appreciate it. This was an awesome conversation. I could go three times as long if I had the bandwidth too. So appreciate the back and forth. Thank you so much for being here today.

Hannah Hunt [00:31:08]:
Yeah, thanks, Ryan. This is awesome. I really appreciate it.