Aerospace and Defense Review

I'm a solo founder with AI agents instead of employees. My 'council' of AI agents saves me 20 hours a week.

• Aaron Sneed created "The Council," a group of 15 custom GPTs that help him run his business.

• The council includes a chief of staff AI agent, a legal AI agent, an HR AI agent, and more.

Stay ahead of the industry with exclusive feature stories on the top companies, expert insights and the latest news delivered straight to your inbox. Subscribe today.

• This article is part of "The AI-Powered Solopreneur," a series exploring how solo business owners use AI to drive growth.

This as-told-to essay is based on a conversation with Aaron Sneed, a 40-year-old defense-tech solo founder based in Florida. The following has been edited for length and clarity.

When I started my business, as a solopreneur, I realized I didn't have the money to pay lawyers, HR reps, and a bunch of other companies. So, using AI, I created what I call 'The Council.'

The council, which is compiled of all AI agents, helps me save around 20 hours a week, and that's a very conservative estimate. Every kind of general corporate chair, HR, legal, and finance AI agent has a seat on the council.

Here's how I use around 15 custom agents, including a chief of staff agent, to manage my workload.

Business Insider wants to talk to founders of AI-powered companies with fewer than 10 workers, and employees working alongside AI agents and bots, to understand what the "Tiny Teams" era really feels like — the wins, the fears, and the human skills that stand out. Share your story by filling out this quick form.

I've been using automated tools for at least a decade

I've worked on autonomous platforms that make decisions independently for at least 10 years. That made me latch onto commercial large language models and AI tools very quickly when they came out.

I primarily use Nvidia's platform as my underlying hardware for technical prototyping and experimentation. I use their GPUs, and they provide free access to their AI software since I purchased their hardware. Additionally, my council is built on OpenAI's ChatGPT business platform using custom GPTs and projects.

Altogether, my AI council consists of the following:

• chief of staff agent

• HR agent

• finance agent

• accounting agent

• legal, comms, and PR agents

• security and compliance agent

• engineering agent

• quality agent

• supply chain agent

• training agent

• manufacturing agent

• business systems agent

• facilities agent

• field operations agent

• IT and data agent

Each agent has a different level of authority

My chief of staff agent is important because it's the voice that sets priority based on parameters like risks, issues, and opportunities.

I told my chief of staff which models have priority when making decisions. For example, anything legal, compliance, or security-related will be given a higher priority. So, I tell the chief of staff to listen to these models over everyone else.

I trained my AI agents to push back and not just say 'yes'

I don't want a bunch of yes agents. I trained them purposefully to give me pushback because I've learned that they naturally want to agree with me. I want them to test my theories to help me with what I'm trying to accomplish.

I have a roundtable set up with all my AI agents, where I can put something like a request-for-proposal document in the chat, and all the agents will weigh in at the same time. I use this roundtable as a level of prevention for hallucinations and knowledge gaps.

The training never really stops, because if I don't continuously train the models, I won't get the outputs I want or need. It takes me about two weeks to train my agents to the level of experience they need to be at for me to feel confident in them. Early on, it took me longer to produce a deliverable than if I'd just done it myself because I hadn't focused properly on training.

I've become a better prompter through training my AI agents

The models have gotten better, and my prompting has, too. I have a better understanding of what information should be in an agent, like having a governance structure for priorities. I have a set of files that put those requirements in place to mitigate the risk of hallucination and false or bad information.

All of the AI companies have different prompt engineering guides. I recommend taking the time to look at them because there's a lot of user error that causes things to slow down when working with AI.

It takes time to get the agents to a good place. A lot of companies are going to try to jump into using AI too quickly for too much without understanding how to use it properly, and these companies could hurt themselves in the long run.

AI replaced roles — but it hasn't replaced human judgment

I'm ill-equipped to handle a lot of these roles and responsibilities, but I'm also forced to do it because I'm bootstrapped.

With my legal agent in particular, I've learned the bounds of putting AI tools into real-world practice. I have a lawyer, and I use my legal agent to try to do some upfront work before handing documents off to my lawyer for a patent or dispute case, or anything like that.

When I was training my model to help me use facts and data to plot a case, I had a lot of information laid out, and I thought what my legal agent created sounded good to me as a non-lawyer. Then I presented all that information to my lawyer, and he said that it was technically and factually correct, but we don't want to express that information because it shows our cards going in.

His legal skillset helped me realize that, even though I thought my agent was correct and ideal to use, it still won't replace a lawyer with that human context, experience, and skills.

Ideally, I would have an HR person, a legal person, and so on, and each would have their own chief of staff AI agent who would help them out. That's what I think the future will look like.

More in News

Aviation welding decisions often begin after the easy option has disappeared. A component arrives from an aircraft system or engine assembly, and replacement stock is unavailable or tied up in a long procurement cycle. The repair order is no longer a routine purchasing task. It becomes a judgment about whether a shop can return a part to service without turning scarcity into scrap. That pressure changes how buyers should read the market. Price matters, but an unusually low quote can be the least useful signal when the work involves exotic alloys, prior repairs, heat sensitivity and limited room for correction. Turnaround matters too, especially when a delayed part holds an aircraft or engine repair in place. The deciding question is narrower. Can the provider understand the part, control the weld process, verify the repair and protect the customer from a preventable write-off? Specialized experience carries more weight in this service than general capacity. Aviation welding does not reward a broad metalworking background unless it is backed by years of aircraft repair exposure and shop judgment. A buyer should look for evidence that technicians have lived with airframe and power plant repair work long enough to recognize where heat, distortion, access constraints or prior repairs can change the job. Training pipelines are slow in this field. A provider that can retain experienced licensed mechanics and certified welders reduces the risk that complex work will be assigned to staff still learning the true limits of the part. Inspection discipline should sit close to the weld bench rather than appear only at final release. Non-destructive testing, final visual review, documented approvals and repair methods tied to manufacturer manuals help buyers separate controlled repair work from improvised salvage. For parts that cannot tolerate a failed attempt, the inspection path needs to be visible before the purchase order is placed. It should also match the repair profile. Some jobs call for penetrant inspection, while others require x-ray review. Skilled visual judgment still matters because aviation weld failures often begin in small signs. Service breadth has value when it removes handoffs that introduce delay or accountability gaps. Welding frequently needs pre-weld or post-weld heat treatment and sheet metal or machining support can affect fit and return-to-service timing. Outsourcing each step may be acceptable for standard work, but one-off aircraft repairs punish loose coordination. Buyers gain more control when related repair stages stay inside the same shop and follow one quality record. Hi Tech Welding is a premier choice because its model is built around the niche work many shops avoid. It’s an FAA-certified repair station focused on aviation welding, airframe and power plant repair, DER service and engineering orders, NDT Level II inspection, heat treatment, sheet metal and machine shop support. Its strongest fit is not routine fabrication. It is the repair of aircraft parts that are difficult to replace and costly to lose. The company’s third-generation aerospace repair background, long-tenured welding staff, in-house support processes and focus on nonreplaceable parts make it a practical choice for buyers who need repair confidence more than broad vendor scale. ...Read more
A grounded piston trainer rarely creates a scheduling problem in isolation. Magneto and accessory repairs sit inside a narrow maintenance window where lead time and technical clarity both affect aircraft availability. A delayed 500-hour inspection can force a school to cancel lessons, move students, buy a replacement unit prematurely or tie up a mechanic in repeated troubleshooting. The invoice may look modest next to an engine event, but the hidden cost often appears in idle aircraft, strained schedules, hurried purchasing and missed utilization. For operators running tight training calendars, the buying question is less about finding any repair source and more about finding one that can protect schedule confidence without hiding weak workmanship behind speed claims. That distinction is where provider comparisons become sharper. Speed still has to be earned. A fast shop that cannot explain inspection findings or parts requirements merely transfers uncertainty back to the operator. Many buyers in aviation have learned that communication is not an administrative extra. It is part of the repair. Operators need access to someone who understands the component, can distinguish a magneto fault from a broader ignition issue and can explain why a replacement part is warranted. When that technical conversation is weak, the aircraft may return to service with the same complaint unresolved. Documentation and pricing also matter, especially for smaller operators that do not have excess aircraft or deep parts inventories. Itemized billing, clear inspection notes and disciplined use of approved manuals give maintenance teams a defensible trail for their own records. That trail reduces disputes and shortens approval cycles. Fair pricing should not mean bargain repair. It should mean that labor, parts, testing and exchange choices are visible enough for the buyer to understand the decision. Repair economics improve when replacement is not the default response to every worn component. The best repair partners also avoid the false tradeoff between pace and care. Extra paperwork does not always correct a weak process. The better test is whether the shop has experienced technicians and clear inspection points, backed by a habit of addressing root causes rather than building delays around them. For piston engine accessories, small errors can travel far. Aircraft knowledge has to sit beside checkoffs, not behind them. Technology belongs in a supporting role. Online payment and faster document exchange can reduce friction, but neither replaces component knowledge. In this field, the most useful modernization is often not a new portal. It is a shorter path between the mechanic and the person who can answer a technical question without passing it through layers of intake staff. Shrike Aero is a premier choice for operators that need magneto and piston engine accessory repair without losing the direct technical relationship that many larger service channels have thinned out. An FAA-certified part 145 repair station, it focuses on Bendix and Slick magneto repair, 500-hour inspections, full overhauls and troubleshooting, with additional capability on selected starters and alternators. Its five-business-day turn target, core bank support, exchange options and itemized pricing address downtime and cost exposure directly. Post-sale troubleshooting and inspection-based service make the recommendation practical rather than promotional. It combines speed with people who know the parts. ...Read more
Space geodetic parameter estimation software system distributors in APAC are seeing a deeper opportunity as reference-frame science, VLBI analysis and satellite co-location research become more important. These applications are highly specialized, but they underpin precise positioning, Earth orientation monitoring and long-term geospatial stability. VLBI remains central to global geodesy because it supports Earth orientation and celestial reference-frame work. NASA’s Space Geodesy Data Analysis Software System is a software suite for analyzing VLBI observations, including processing raw interferometry visibility data and supporting geodetic analysis.  This creates a narrow but critical software-distribution category. Users working with VLBI data often need more than a downloadable package. They need assistance with data formats, model selection, parameter estimation strategy and interpretation of residuals. Distributors serving APAC institutions must be able to support both scientific workflows and local technical capacity building. There is also the development of new software. A 2026 paper presented GASV, which is a Python-based package for analysis of VLBI in geodesy and astrometry. It can perform both pipeline and interactive processing, and it estimates station positions, EOPs, source positions, clock parameters and atmospheric models. The researchers found results comparable to those from analysis centers in BKG and USNO on selected sessions. This presents a market opportunity for distributors because traditional geodetic software can be very useful, but it can be very hard to install, configure and maintain. While new software that simplifies the workflow process may increase the number of users from just experts to others, validation should be done first before using it in production processes. The APAC region also has VLBI observation infrastructure. A 2026 study showed comprehensive VLBI observations of Galileo satellites with the Australian AuScope array using the antennas in Hobart, Katherine and Yarragadee. This shows the feasibility of future co-location satellite missions. It provides groundwork for future missions like Genesis by ESA. This type of research expands software requirements. Processing VLBI observations to navigation satellites is not standard in every workflow. It requires correlation, fringe fitting, precision assessment and parameter estimation that can connect VLBI and GNSS frames. Specialist distributors can help institutions evaluate whether their software systems support these newer use cases. Reference-frame stability also affects practical applications. Surveying, mapping, sea-level monitoring, satellite orbit determination and disaster-risk analysis all depend on consistent geodetic foundations. When software errors or outdated models enter the chain, downstream users may not see the problem immediately. The challenge is market size. VLBI and high-end space geodesy software distribution is not a high-volume segment. It requires long sales cycles, institutional relationships and deep technical credibility. Providers may need to combine software distribution with consulting, training and managed processing services. The next phases of the market will likely favor distributors that can bridge research tools and operational geodesy. APAC institutions need access to advanced software, but they also need support that makes the tools dependable in national and scientific workflows. Space geodetic parameter estimation software system distributors in APAC are becoming reference-frame support partners. Their value will be measured by whether they help institutions maintain precise, modern and interoperable geodetic analysis capability. ...Read more
Space geodetic parameter estimation software system distributors in APAC are being reshaped by the growth of multi-GNSS and real-time positioning. Users are no longer working only with one satellite system or delayed post-processing results. They increasingly need software environments that can handle multiple constellations, real-time streams and advanced correction products. The International GNSS Service describes itself as a service of the International Association of Geodesy, the Global Geodetic Observing System, the International Union of Geodesy and Geophysics and the International Science Council World Data System. It provides openly available high-precision GNSS data and products for scientific and operational use.  This is important in the context of APAC as the region uses a number of different satellite constellations and augmentation systems. The end users might have to use GPS, Galileo, BeiDou, QZSS and regional data streams according to their applications and regions. The software needs to be able to handle the observation types, biases, orbit and timing standards of different systems. The high-accuracy positioning is gaining importance as a market-driving factor. A study on the GNSS market up to 2026 points out that the real-time kinematic and precise point positioning are driving the use of GNSS technology, with the ability to provide accuracy beyond the consumer navigation requirements. This increases demand for parameter estimation tools that can support centimeter-level workflows when field and data conditions allow. Distributors have a key role in implementation. A customer may purchase software for PPP, orbit determination or station coordinate estimation, but performance depends on correct configuration, reference products and processing strategy. Poor setup can produce results that appear precise but are not reliable. Software capability is also evolving. GipsyX/RTGx, developed at JPL, is described as a tool set for positioning, navigation, timing and Earth science using GNSS, SLR and DORIS, with VLBI under development. It can estimate station coordinates, satellite orbits, clocks, Earth orientation and atmospheric delays in post-processing and real-time contexts. This type of capability raises the support burden. APAC distributors must be able to explain Kalman filtering workflows, reference-frame assumptions, data quality constraints and output validation. They may also need to support integration with GIS, monitoring dashboards or national positioning services. Industrial cooperation is also becoming important. GNSS.asia says it facilitates industrial cooperation on GNSS between Europe and the Asia-Pacific and has supported more than 80 businesses since 2012. This shows that distribution in the region is not only a sales activity. It is also part of technology transfer and ecosystem building. The challenge is user diversity. A university research group, a national geodetic agency and a precision agriculture technology firm may all use GNSS data differently. Software distributors need flexible training and support models for each audience. The next phases of APAC geodetic software distribution will likely favor companies that combine product access with applied positioning expertise. Customers need tools, but they also need confidence in the processing chain. Space geodetic parameter estimation software system distributors in APAC are becoming real-time positioning enablers. Their strongest value will come from helping users manage multi-GNSS complexity while preserving accuracy, traceability and operational reliability. ...Read more