To ensure that I answered the mail to the handful of people that contacted me asking about the fiction focused version of the workflow I shared yesterday, it is shown below. I'm going to devote some time this weekend to cleaning both up into full guides with step-by-step templates, so look for those updates on my website soon.

Just when I thought I had finally neared completion of editing my book, I also finally achieved full success on my GenAI manuscript editing workflow, and damn did it work. It’s both an amusing moment to solve a series of problems that unlock a huge capability, but also sort of frustrating to realize how much writing/rewriting I still have left in my book. The results of both of those situations are ultimately very positive for the book and all my future GenAI workflows. For context, my previous editorial workflows were generating maybe 1-2 pages of notes, that were mainly just resulting in positive reflections of the current content with a few suggestions for content improvement. The new workflow breakthrough yielded over 8 pages of extremely detailed and high-quality feedback that was fantastic! It really was one of those magical “AH-HA” moments of both achievement and realization.

This is the one of those practical hardware engineering applications that really gets me excited. We have seen much higher end exoskeleton systems being tested by SOCOM previously, but the trickle down of technology into the commercial market is still nonetheless exciting.

There are usually multiple ways to accomplish the same task, as well as there’s a large amount of preexisting information on the many and optimized ways to analyze data. Are you using the best productivity and/or analytical frameworks for your tasks? Do you know what those options are? Are you choosing and alternating frameworks for each task to optimize results for the specific task?

Thinking back seven years ago to when I was helping run the Thunderdrone event at SOFWERX, there’s a number of interesting lessons to be learned for how the sUAS market evolved from that point, and how those lessons might apply now to the newly growing GenAI market. “We can’t even talk about testing these things with weapons on them”, was the cry from so many DoD leaders at the time (ISR was the primary focus). Security and safety was the top priority regardless of the weapons potential. “We need better, bigger, and more capable platforms”, was the constant drum beat of requirements managers at the time (Group 3+ was the focus for more payload and flight time). Looking forward to today’s battlefield, we see quite the opposite.

It’s a lived technical reality for a few, but not necessarily an obvious realization for many GenAI observers, that the current state of the technology limits the model’s interaction with data to complete interactions only. In simple terms, the models cannot currently interact outside of themselves with separate applications to affect pieces or parts of large data files or applications. A basic example would be the models could perform a grammar check on a document, but not automatically interact with MS Word to fix only the errors it identifies. You would either need to have the model attempt to rewrite the entire (or some whole section of) text to fix the errors (hoping to note induce more in the process), or you would have to manually fix the errors separately yourself. We saw this same thing with diffusion models that generate images until recently, when notable major diffusion model providers started adding additional tools to their user interfaces to allow image subsection alterations. Soon with agents we will see similar capabilities come to LLMs.

I have been excited with all the ways that GenAI tools have been able to help me edit my current book, and it excites me about what more I could do. I really want to go back revisit the concept I wrote about in a post last year regarding using video interviews as a source of analysis for insights. Specifically I was thinking about writing a book or maybe a short series of articles on the evolution of battlefield technology in the Ukraine war. There is ample content in the form of online interviews on this topic, from the front lines to the development labs. Not that there is a shortage of existing written content that discusses this topic, but that’s mostly the point. To see how fast it is possible to pull together meaningful long form content from a larger source repository of audio-visual content.

Anyone frustrated why their SBIR/STTR proposal rejection letter was limited on detailed feedback? Maybe you had multiple similar submissions with differing evaluation results. All can seem a bit confusing when taken as a whole, especially if it’s your first time playing the SBIR/STTR game. Allow me to shed some light on these otherwise murky waters.

There was much gnashing of teeth recently at an early look into the upcoming DAF GenAI system approval process, but instead of complaining, there’s some tricks to system design you can use to help improve your odds of approval. First, one of the biggest issues with any GenAI system is the ethics considerations. Drawing the line on what is and isn’t allowed in or out of the model for use is a very tricky thing. Ultimately some line must be drawn by someone, but therein lies the issue. How does the system owner enforce that limitation?

As of today, we now have open-source LLMs that are at least a 10th as small, but approximately as capable as the original release of GPT4 last year. So, in roughly a year we have seen development in GenAI progress to the point where the baseline standard for open-source LLMs (Meta’s Llama models) catches up to the original groundbreaking foundational model standard of previous year. Now the interesting thing is that this achievement isn’t actually new. We saw the exact same thing last year with the release of Llama 2, as it was roughly equivalent to GPT 3.5 but smaller and released roughly a year after as well. So, this is just more validation of the existing developmental trendline within the field. But the interesting thing isn’t where we are today on that trendline, but where we are all going very soon.

Figured I’d share some interesting points from recent conversations I have had with friends, clients, and partners. Some of these things you might agree with, but I have absolute confidence someone or many will disagree.

It seems like a relatively simple question, but this one fact radically defines your target customer base. Now how big that potential customer base is also happens to define a big chunk of the difficulty of making a sale within the DoD as well. You don’t have to have a niche product, if that product is only purchased by a small if not singular organization within the DoD. The statistical odds of a sale are roughly the same. That isn’t to say those odds are zero, far from it, but if you only have a small number of potential customers that is lower odds than having many and separate possible ones. The point being to understand the game that you are playing. That way you can decide if it is a game you want to play, and then if so, to understand what realistic odds you have of winning that game.

Regardless of your belief that the scaling of Generative AI development will ever achieve a particular extreme computational performance level, one thing that everyone should agree on is that many companies will keep voraciously pursuing more compute resources to at least keep trying. That rapidly scaling compute is extremely power intensive, and data centers can’t just go anywhere due to the limitations of power transmission with current material science. Those data centers also require very stable baseload power at very large capacity, often approaching multiple megawatts or more. Even if you could get around all the environmental and political issues with fossil fuel-based power generation, it’s still bound by the geographic constraints of the natural resource sources that make placing the power plants tricky. The only option is nuclear power for high output capacity, stable, widely deployable power generation.

The Fiscal Year (FY) 2025 Federal budget is likely to be the most drawn-out budgetary process in recent memory. I have discussed this probable reality with several companies in recent weeks, and want to make a more public reminder for everyone else. FY25 will kick off with a Continuing Resolution (CR) like every other FY in the last decade, but it is also an election year. The odds of the presidential election being close if not contested even more than we have seen in last cycle is likely. That will be the highest priority for congressional action overall. Plus, you have potential shifts in the parties as the majority status of either is likely to be and remain razor thin. Thin majorities give party outliers opportunities to be king makers on any number of issues, and we have seen this very thing play out over the few years. I don’t particularly care about politics, but it is important to highlight that the extended political infighting that is very likely on its way in FY25, and will ultimately make next fiscal year operate under CRs for most of the FY. But why does that matter to me, and why am I even talking about this to you?

The DoD has plenty of internship programs, active service volunteer efforts, and myriad of other methods to bring together a team of personnel to pursue prompt engineering on any number of existing platforms. In fact, this has already been done by DAF's CAITO last year in a small-scale effort to produce the first version of a prompt engineering handbook. The core issue is it could and should be an enduring effort focused on personnel training through direct experience. What better way to train the force than through direct experience of rotating participants. Any platform or model can be used, any data or use case could be utilized, and anyone could be a participant. Prompts are the ubiquitous and common tool all models utilize for function, and they can be continuously developed, tested, and cataloged for the evolution of capability across the force.

Simply put, you need to build up your experience around using individual prompts as building blocks to construct the larger effects which you then orchestrate into complete workflows to produce high quality outputs. You do not need private models, specially fine tuned models, or super secure 3rd party services to do any of this type of development for your organization. You could literally have an unpaid college intern do this for you over a few weeks or months with nothing more than a $20/month ChatGPT or Claude account. But what does this development look like?

This is not a shameless plug, but rather just an obvious statement that hopefully is redundant to most defense businesses. All services have at least one, and often multiple directly sponsored and contractually integrated innovation hubs that service them. There are big-name brands that many people know such as: AFWERX, AAL, SOFWERX, NAVELX, etc. There are many more focused or smaller hubs like ERDXWERX, FLEETWERX, ARCWERX, etc. There is a very expansive list of all of these and the direct links to each on my website (https://lnkd.in/gWdUUGCi). What all the innovation hubs share in common, is that they all serve a military organization of some size directly. This means they have direct access and knowledge of customer needs and priorities to varying levels…but all to a greater degree than any external business (short of major defense contractors).

But to keep things interesting let’s just make the answer simple with an easy rule of thumb. If you have a potential Government or Military customer that is willing to pay now, and time is of the essence in any sense (especially for the customer), then go OTA, otherwise go SBIR. Is it that simple, no, but can it be that simple, yes. There are many things to consider, and you could come up with all kinds of justifications for either contract given any number of situations. At the end of the day, people like simple answers to hard questions. To that end, it is best to look to the primary advantages and disadvantages of each to form a rule of thumb.

Just like apples and oranges, they are both fruits that come from a tree, and eventually must be picked by someone to consume, but that’s where the similarities end. Specific Topics are the original form of the SBIR/STTR program dating back decades. The entire plan for all SBIR/STTR phases is mapped out from the start by the Gov/Mil customer with the need. It truly is the unique nail searching for its specific hammer. Open topics on the other hand, are something very different. The Open Topics are in many senses the inverse of Specific Topics, where the hammer goes searching for its applicable nail. The core idea behind Open Topics is that there exist products that could provide a capability to Government or Military users that those potential users don’t know they need, and this program allows businesses to pitch these products to the Gov/Mil. When starting from the solution and working backwards to the need, it’s easy to understand why it is much more difficult to find product market fit, let alone a single customer with all the necessary funding to fully build out and/or acquire the product.

The dirty little secret in military innovation circles is that MANY military personnel regard the SBIR/STTR program as “free money”. This isn’t an entirely inaccurate belief, in that all SBIR programs have some kind of funding support possible to Phase 1 and/or Phase 2 contracts, but the whole point of the program is to get to Phase 3 where it’s all on the customer to fund everything. Rather the customer went any number of other contracting routes to acquire the product or service, or chose to use a SBIR/STTR contract, it all ends the same way with a military customer funding the purchase. The hot potato game of trying to make some other military organization pay for your military organization’s wants or needs, can be the most frustrating and opaque game that many defense businesses end up suffering through.