Hi friends! I have a brand new episode live and I’m so excited to chat with Mark Ennis all about healing your relationship with food, overcoming sugar addiction, and finding clarity through the carnivore lifestyle.
Here’s what we discuss:
– How Mark discovered the carnivore diet and why it transformed his physical and mental health
– The role of routine, mindset, and accountability in achieving long-term health goals
– Common nutrition myths and how women’s dietary needs are often misunderstood
– The connection between food, mental clarity, and emotional resilience
Plus practical tips, encouragement for experimenting with what works for your body, and so much more!
187: Food Freedom & Clarity with Carnivore Coach Mark Ennis
About Mark Ennis: Mark Ennis is a health coach from Ireland specializing in helping individuals achieve a longer, healthier life through the power of a carnivore and keto lifestyle. After struggling with both anorexia and obesity, Mark discovered the life-changing benefits of this approach and now dedicates himself to guiding others on their journey to health and happiness. Passionate about making a lasting impact, he empowers clients to overcome challenges, achieve their goals, and live their best lives. Mark’s mission is to inspire positive change and share the knowledge that transformed his own life.
If any of my fellow health professional friends are looking for another way to help their clients, I highly recommend IHP. You can also use this information to heal yourself and then go one to heal others, which I think is a beautiful mission. You can absolutely join if you don’t currently work in the health or fitness industry; many IHPs don’t begin on this path. They’re friends who are passionate to learn more about health and wellness, and want to share this information with those they love. You can do this as a passion, or start an entirely new career.
Thank you so much for listening and for all of your support with the podcast! Please be sure to subscribe, and leave a rating or review if you enjoyed this episode. If you leave a rating, head to this page and you’ll get a little “thank you” gift from me to you.
Success! Check your email for a free 30-day meal and fitness cheat sheet
Let’s be honest: B2B marketing is facing a trust crisis. Our ebooks and webinars are now met with skepticism. Buyers are drowning in a sea of AI-generated noise and unsupported claims.
As property/casualty insurers increase their focus on predicting and preventing costly damage that drives up claims and premiums, telematics technology has come to play an increasing role. From video doorbells that reduce theft and vandalism to “smart plumbing” solutions that detect leaks and shut off water before in-home flooding can occur, these technologies clearly offer value to homeowners and insurers.
But how much value?
Whisker Labs – maker of the Ting home fire prevention solution – has taken on the challenge of quantifying its product’s efficacy and return on investment. In a research partnership with Octagram Analytics for independent data analysis and modeling and Triple-I for its insurance industry expertise and insight, Whisker Labs found that Ting reduced fire claims within the study sample by an estimated 63 percent, resulting in 0.39 fewer electrical fire claims per 1,000 home years of experience, in the third year after installation. This translates into a fire claims reduction benefit of $81 per customer.
“This study provides concrete evidence of the value that telematics technology can deliver,” said Patrick Schmid, chief insurance officer at Triple-I. “While IoT solutions are gaining traction with many success stories, rigorous analysis of claims reduction has been harder to find until now. This analysis clearly shows Ting reduces claims and provides a positive return on investment for insurers.”
Ting helps protect homes from electrical fires by using advanced AI to detect arcing, the precursor to most electrical fires. Once connected to a single outlet, Ting analyzes 30 million measurements per second, analyzing voltage at high frequencies to detect tiny electrical anomalies and power quality problems. These hazards can originate from wiring in the home, connected devices and appliances, or even the power coming in from the utility. On average, Ting detects and mitigates fire hazards in 1 out of every 60 homes it protects.
“Ting is about saving lives and homes – that’s always been our mission,” said Bob Marshall, CEO and cofounder of Whisker Labs. “By analyzing verified claims data over time, this analysis shows that what’s best for families also delivers a strong financial return for insurers. Prevention is better for everyone.”
Whisker Labs works with a growing community of 30 insurers who provide Ting to their customers for free. More than one million Tings are deployed in the United States, and approximately 50,000 new Tings are installed each month.
In addition to monitoring voltage and features of voltage at high frequencies to detect arcing that is indicative of fire hazards, Ting has a temperature sensor that monitors the temperature within the home.
“When the temperature drops below 42 degrees, an alert is issued,” Marshall said. “Thus, Ting detects and warns about conditions that can result in frozen and burst pipes and alerts the homeowner to correct the situation before damage occurs. Over the past three years, we have issued low-temperature warnings to about 1 in 560 customers per year.”
Measuring the value
Like Ting, other peril-based IoT solutions issue alerts and warnings when a hazard is detected. Thousands of hazards are detected and alerts sent, but how do you know that this reduces claims? How do you estimate the return on investment for these devices? How can you prove that the bad thing, a loss and a claim, didn’t occur?
“We developed a methodology to do this in the real world with existing customers and experience data,” said Whisker Labs Chief Scientist Stan Heckman.
Whisker Labs and Octagram had to overcome challenges related to limited data and sampling bias. To address these, a self-controlled study was developed that assesses claims over time in homes with Ting in place. (See paper for a fuller explanation of the methodology).
The chart below shows how the number of fire claims in Ting-equipped homes declines over time. The claims frequencies observed and associated percent reduction in claims are highly dependent on the definition of the sample of non-cat fire claims provided by carriers that participated in the analysis. However, this does not affect the observed absolute reduction.
Using data from Triple-I and Verisk, Whisker Labs determined that Ting provides a loss-prevention benefit of $81 per home per year. (See paper for details).
“Add in benefits associated with reduction in water-related losses from frozen pipes and failing sump pumps and water heaters,” and the benefits are likely substantially higher, Marshall said. Insurers who provide Ting to their policyholders also may enjoy improvements in customer retention.
Budgeting apps are everywhere, and for good reason. They promise to help you track spending, reduce waste, and gain control over your finances. But what most users don’t realize is that many of these tools have a business model that doesn’t rely on your subscription fee. It relies on you. Or more specifically, your data.
From your purchase habits to your financial goals, budget apps collect a surprisingly intimate profile of your life. And in many cases, they turn around and sell that information to advertisers, data brokers, or even financial institutions looking to target you.
1. Mint: Convenience at the Cost of Privacy
Once a leader in the budget app world, Mint was a go-to for millions of users until it shut down in 2024 and urged users to switch to Credit Karma, another Intuit-owned platform. But even before that, privacy experts had flagged Mint for its data practices.
Mint links directly to your bank accounts, categorizing your transactions with impressive ease. But all that data fed a much larger machine. Intuit used the information to power marketing strategies, product development, and data sales. Many users didn’t realize that their budgeting behavior was helping Intuit sell them financial products, like loans, credit cards, and insurance.
2. Rocket Money: Your Subscriptions Aren’t the Only Thing Being Tracked
Rocket Money (formerly Truebill) gained popularity by offering to cancel unused subscriptions, track spending, and even negotiate bills. But buried in its privacy policy is language that allows the app to collect and share information with third parties for “marketing purposes.”
This includes transaction data, bill payment history, and your interactions with partner services. Some users have reported an uptick in financial product ads shortly after linking their bank accounts to Rocket Money—a red flag that suggests your data is doing more than just budgeting behind the scenes.
3. EveryDollar: Clean Interface, Cloudy Data Practices
Created by the Ramsey Solutions brand, EveryDollar is marketed as a faith-based, common-sense budgeting solution. While its paid version offers more robust features and less data sharing, the free version may still collect data that can be used for marketing.
Because the app’s terms of service allow for the collection of “non-personal” usage data, you could still be feeding a profile that advertisers can use to find people just like you. Even anonymized data has become valuable to marketers who want to target by income range, financial goals, or spending behaviors.
4. Goodbudget: Envelope Budgeting with a Side of Tracking
Goodbudget is based on the traditional envelope method—assigning portions of your income to specific spending categories. It’s a useful tool for discipline-focused savers, and it has a loyal following.
But even simple apps have fine print. While Goodbudget isn’t as aggressive in data sales as some others, it still collects metadata, app usage behavior, and device information. That may seem harmless—until you realize how easily it can be used to infer income, lifestyle, and even political leanings based on spending patterns. And because it integrates with your browser and email for syncing, your data footprint might be larger than expected.
5. Simplifi by Quicken: Premium Pricing Doesn’t Guarantee Privacy
You’d think a paid app like Simplifi would have less incentive to sell your data. After all, you’re the paying customer—shouldn’t that mean your data is safe?
Unfortunately, that’s not always true. Simplifi’s terms of service allow the use of financial data for internal marketing and product development. While they claim not to sell personally identifiable information, the app may still use your behavior to fine-tune ads or promotions from third-party partners. This creates a murky line between “internal use” and profiling that benefits advertisers.
6. YNAB (You Need A Budget): Better Than Most, But Not Perfect
YNAB is often praised for its privacy-first approach. It doesn’t sell data, its business model is based on subscriptions, and it has clear, user-friendly privacy policies. But even here, there are a few things to be aware of.
Like most digital tools, YNAB uses cookies and third-party services for analytics. This includes Google Analytics and marketing pixels that track how you interact with the site and app. While this is less invasive than financial data sharing, it still contributes to an advertising ecosystem where your behavior is observed and potentially monetized.
Why These Apps Are So Hungry for Data
Budgeting apps, especially free ones, often rely on data monetization to stay afloat. This includes selling:
Spending trends in banks and retailers
Credit behavior to financial advertisers
Demographic targeting for political campaigns or insurance companies
Because financial data is especially revealing, it’s incredibly valuable. Advertisers will pay a premium to reach someone who’s actively budgeting, managing debt, or saving for a home. The apps don’t have to sell your name. They just have to sell access to “people like you.” And once you’ve linked your bank accounts or credit cards, the app has a complete picture of your habits, priorities, and struggles.
What You Can Do to Protect Yourself
If you’re already using one of these apps, don’t panic, but don’t stay passive either. Here are a few ways to take control:
Read the privacy policy (yes, the whole thing)
Opt out of marketing and data sharing when possible
Use anonymized or read-only versions of your financial data where available
Consider paid services with strong privacy commitments
Regularly review your app permissions in both iOS and Android
You might also consider budgeting the old-fashioned way, using a spreadsheet or an offline tool. No app can sell what it doesn’t know.
Free Budgeting Isn’t Really Free
In a world where your personal data is more valuable than ever, free budgeting apps often come with invisible price tags. From selling your spending habits to profiling your lifestyle, these tools can compromise your financial privacy in subtle but significant ways.
The apps listed here aren’t inherently bad, but they’re part of a digital economy that profits from your behavior. And the more financially vulnerable or budget-conscious you are, the more appealing your data becomes to advertisers.
Have you ever caught an app sharing more than it should? What’s your go-to budgeting method now?
I distinctly remember my first hot flash. My husband and I had stopped for breakfast on our way to Atlantic City for an overnight stay. Sitting in the booth across from him, I suddenly felt heat rising through my body—like someone had turned up my internal thermostat. I started laughing. “I think I’m having my first hot flash,” I said. Moments later, I stood up and walked outside. It was as if I had heated up the air around me and needed to move.
That was more than six years ago.
The First Hot Flash—and the Perimenopause Symptoms That Followed
For me, perimenopause started with hot flashes. During the day, they struck mostly while I was eating. In mid-winter, I would jump up from the table to go stand outside. As someone who has been perpetually cold, it was odd.
Then came the night sweats—waking up with my PJs so drenched that I would change my clothes, only to wake up drenched again a couple of hours later.
Next was the insomnia. I had always slept like a champ, falling asleep almost as soon as my head hit the pillow. I could still fall asleep easily (most nights), but I would wake up at 2 a.m. It was as if my body and brain were on two different rhythms. My brain and nervous system were tired, but my body was so awake. I swear I could actually feel the pulses of energy running through my legs. Sometimes I would get out of bed. Sometimes I was so tired I would just lie there feigning sleep.
I could handle the occasional sleepless night, but sometimes several of these nights would occur back-to-back. By the third or fourth day, I felt like garbage and my brain was mush.
My Symptoms Were Brushed Off for Years
I mentioned the night sweats and insomnia to both my primary care physician and my nurse midwife at each visit for years. Each time, they assured me it was all “normal” and “part of the transition.” I trusted them. And—on some level—I think we all assumed these signs were an indication that menopause was around the corner and, therefore, these symptoms would come to an end anytime now. But they didn’t. They got worse.
I mean, the average woman reaches menopause (officially: the day that you’ve gone a full 12 consecutive months without a menstrual period) at age 51, and I’m 54 now. It made sense.
Only, menopause wasn’t around the corner. I still get my period like clockwork.
Perimenopause Brought Me A Miserable Monthly Cycle
In the last two years, my perimenopausal symptoms got worse and worse. And while they didn’t all occur every month, I started keeping notes and realized that most of them occurred cyclically, often hitting mid-cycle. Things like:
Painful ovulation and period cramps
Constipation
Bloating
Red, swollen, bleeding gums
Mouth sores
Vulvar swelling and irritation
Sore, swollen breasts
Low libido
Days when I’d randomly wake up feeling anxious, sad, or pissed off
In short, I was miserable. The sleepless nights and revolving door of symptoms made most days feel like a complete slog. I just didn’t feel like myself. (Little did I know how common this is!)
Finally Finding Help
Desperate, I found a new gynecologist and went to my visit with notes, prepared to discuss my symptoms and determined to ask if hormone replacement therapy was an option. I wasn’t halfway through my list before he dismissed most of the symptoms. I didn’t push—who wants to work with a doctor who gaslights them?
After the appointment, I sat in my car and cried. I was so frustrated. And I felt like I was back at square one.
I’m not alone.
The Medical System’s Menopause Gap
Despite the fact that nearly 90 million women in the U.S. are expected to be postmenopausal by 2060, menopause remains a profoundly underserved area in medicine.
Most women will spend about one-third of their lives in this stage, yet both patients and providers are often unprepared for what it brings.
Research shows that while 85% of women experience menopausal symptoms that significantly impact their quality of life, a staggering 75% of those who seek help walk away untreated.
Meanwhile, only 54% of women can accurately define menopause, and 32% say they lack basic knowledge about it.
Unfortunately, the providers they turn to may not be much better equipped: 80% of internal medicine residents report feeling unprepared to treat menopause, and only 20–30% of OB/GYN residencies include formal menopause education.
Read that again: Only 20–30% of those who go to medical school to work with people who have uteruses are formally educated in menopause!
Much of the confusion around hormone replacement therapy (HRT)* can be traced to the 2002 release of the Women’s Health Initiative (WHI)—a large, government-sponsored study that linked HRT to increased risks of breast cancer, stroke, and heart disease.
*Editor’s Note: Menopause hormone therapy (MHT) or simply hormone therapy (HT) are the currently-accepted terms from the Menopause Society and the Endocrine Society for the practice of prescribing hormones related to menopausal symptoms. As pointed out by Dr. Jen Gunter here in her Substack, The Vajenda, use of the terms “Hormone Replacement Therapy or HRT implies that menopausal women have a disease.” We highly recommend every midlife woman subscribe to Dr. Gunter’s Substack.
The findings were broadcast widely and prompted millions of women to discontinue HRT, while physicians were advised to prescribe it sparingly. What many headlines left out, however, was the nuance: the average participant in the WHI was 63 years old, over a decade past the average age of menopause. Most had pre-existing health conditions, and the hormones used in the study—oral Premarin and synthetic medroxyprogesterone—are now known to carry higher risks than the currently prescribed bioidentical hormones delivered transdermally.
In the years since, follow-up studies and re-analyses have shown that when started closer to the onset of menopause, HRT—especially formulations using bioidentical estradiol and micronized progesterone—can be not only safe but beneficial, improving quality of life and reducing risk of cardiovascular disease and osteoporosis. Yet the stigma and misinformation from the early 2000s continue to influence medical guidelines, media narratives, and public opinion.
Finding the Right Doctor is Key
The weekend after my disastrous appointment, I met a couple of friends for breakfast. I filled them in on my worsening symptoms, my frustration, and the disappointing visit with the new doctor. One of my friends—Jen—had been just as frustrated with her perimenopausal symptoms and told me she’d found a local practitioner on a list of recommended practitioners. She had already scheduled an appointment with Dr. Mary Ann Yehl and would share her thoughts after.
“Gals: Two thumbs up for this menopause doc I saw today,” Jen texted after her appointment.
That was all I needed to hear. I promptly scheduled an appointment with Dr. Yehl.
Afterward, Jen and I talked more about our mutual frustrations navigating perimenopause in a broken healthcare system. “Between the two of us, we had to cycle through six doctors just to find one who had the knowledge—and the willingness—to talk about what we knew was happening to our bodies,” she told me. “One doctor literally said, ‘We don’t give hormones just so someone can sleep.’ That might be the most enraging thing I’ve ever heard from a doctor.”
What stuck with her most was the inequity of it all:
“I’m lucky to have a supportive partner and the means to afford out-of-network care—but I kept thinking about all the women who won’t get the help they need because they don’t have the same privilege. I was relieved to find Dr. Yehl—but also furious. It just shouldn’t be this hard to get basic care.”
The Appointment
My appointment with Dr. Yehl was everything a doctor’s visit should be. She asked me to tell her my story—and then truly listened. She didn’t just focus on the obvious symptoms; she asked about diet, exercise, and my emotional, mental, and physical health in equal measure. The validation she offered was priceless. And she was so thorough that she caught something both my primary care doctor and previous gynecologist had missed: uterine fibroids. (But that’s a story for another post.)
Two months into hormone replacement therapy, I could cross off every single symptom on my list. No more sleepless nights. No more constipation. No more mouth sores or bleeding gums. No more feeling like I was unraveling mid-cycle.
As Dr. Yehl later shared with me:
“In an ideal world, we’d sit down with women around age 35 to give them a preview of perimenopause and menopause—what symptoms to look out for and how to prevent chronic disease. So many women feel like they’re losing themselves as anxiety, depression, cognitive changes, and physical symptoms creep in. It’s a very isolating time for many women when their bodies and minds change in unsettling ways. If they only knew that there are doctors who understand—and can recognize and treat these symptoms. There is hope, and there is help.”
The Cost of Going to a Menopause Specialist
Most menopause specialists are out-of-network for health insurance. The few practices I reached out to charged anywhere from $400 to $975 for an initial visit—though things are admittedly pricey in the NY/NJ area.
Add to this the fact that most health insurance companies only cover some forms of estradiol and often don’t cover progesterone or testosterone at all. At present, I shell out $67 a month for HRT.
Many women don’t have the resources to jump through the hoops necessary to get the care all women should have access to. That’s the part that’s hardest to swallow.
You Are Not Alone
If you’re struggling with perimenopausal or menopausal symptoms and feeling dismissed or confused, you’re not alone—and you don’t have to suffer in silence. The lack of menopause education in our healthcare system leaves too many women untreated and unheard. But there are providers who are knowledgeable, compassionate, and ready to help. To find a menopause-informed practitioner near you, visit The Menopause Society.
How has your perimenopause journey been going?—Karin
Lobo, R. A. (2005). WHI clinical trial revisit: Imprecise scientific methodology disqualifies the study’s outcomes. American Journal of Obstetrics and Gynecology, 193(4), 1030–1036. https://www.ajog.org/article/S0002-9378(05)01186-5/fulltext
PubMed. (2023). Needs assessment of menopause education in United States obstetrics and gynecology residency training programs. Menopause, 30(10), 1150–1158. https://pubmed.ncbi.nlm.nih.gov/37738034/
Last week, former White House Communications Director Anthony Scaramucci offered a surprising assessment of Donald Trump, calling his political comeback historic and cautioning critics not to confuse his communication style with a lack of intelligence.
What Happened: In a video posted to his YouTube channel, Scaramucci—who once described working for Trump as the “biggest mistake” of his life—expressed his stance on the president’s intellect and political savvy.
“If you think the president is stupid, you don’t know the president,” Scaramucci said. “He may talk like a fifth grader … but he’s a very smart and very clever guy.”
Scaramucci, who briefly served as Trump’s White House communications director in 2017, suggested Trump may struggle with conditions like dyslexia or ADHD but compensates with heightened situational awareness.
He also praised Trump’s political strategy, saying, “He understands this trick—the trick is to tell the big lie and to repeat the big lie and to create a mantra and narrative around it.”
Calling Trump’s return to the White House “the greatest comeback in political history,” Scaramucci noted, “He went from being a reality television star and a moderately successful real estate developer to the American presidency in 18 months—and he just reascended to the presidency.”
Why It’s Important: While Scaramucci has commended Trump for advancing digital asset regulation and establishing a Strategic Bitcoin Reserve without taxpayer funding, he has also criticized several aspects of Trump’s presidency, including tariffs and economic decisions.
He strongly condemned the launch of Trump’s official meme coin, the Official TrumpTRUMP, calling it “Idi Amin-level corruption” and likening a related Truth Social post to a market “pump.”
Trending Investment Opportunities
Scaramucci also joined others in questioning the timing of Trump’s crypto-related announcements, suggesting possible manipulation.
Despite his political criticisms, Scaramucci holds substantial crypto investments, with over 55% of his net worth in BitcoinBTC/USD and a significant stake in SolanaSOL/USD.
Earlier this year, Scaramucci also said that his brief 11-day stint in the White House during Trump’s first term rekindled his interest in Bitcoin.
A July 2025 report by Fortune said that, as per sources, Scaramucci’s net worth is estimated to be between $150 million and $200 million, largely tied to volatile crypto assets.
His wealth stems from personal investments and management fees through SkyBridge Capital, which had $2.6 billion in assets at the end of 2024. He also earns royalties from several books, including The Little Book of Hedge Funds.
While some in the crypto industry criticize his financial expertise, others, like SALT CEO John Darsie, credit Scaramucci for strategic pivots that saved SkyBridge—such as shifting from hedge-fund seeding to a fund-of-funds model via a Citi acquisition in 2010 and moving heavily into crypto in 2020.
Earlier this week, Scaramucci also voiced support for Trump’s infrastructure-focused spending—if it creates long-term economic value—calling for “Big, Beautiful Spending” with a real multiplier effect, like Boston’s “Big Dig.”
Read Next:
Photo Courtesy: Al Teich On Shutterstock.com
Disclaimer: This content was partially produced with the help of AI tools and was reviewed and published by Benzinga editors.
Hey there, everyone, and welcome to the latest installment of “Hank shares his AI journey.” 🙂 Artificial Intelligence (AI) continues to be all the rage, and coming back from Cisco Live in San Diego, I was excited to dive into the world of agentic AI.
With announcements like Cisco’s own agentic AI solution, AI Canvas, as well as discussions with partners and other engineers about this next phase of AI possibilities, my curiosity was piqued: What does this all mean for us network engineers? Moreover, how can we start to experiment and learn about agentic AI?
I began my exploration of the topic of agentic AI, reading and watching a wide range of content to gain a deeper understanding of the subject. I won’t delve into a detailed definition in this blog, but here are the basics of how I think about it:
Agentic AI is a vision for a world where AI doesn’t just answer questions we ask, but it begins to work more independently. Driven by the goals we set, and utilizing access to tools and systems we provide, an agentic AI solution can monitor the current state of the network and take actions to ensure our network operates exactly as intended.
Sounds pretty darn futuristic, right? Let’s dive into the technical aspects of how it works—roll up your sleeves, get into the lab, and let’s learn some new things.
What are AI “tools?”
The first thing I wanted to explore and better understand was the concept of “tools” within this agentic framework. As you may recall, the LLM (large language model) that powers AI systems is essentially an algorithm trained on vast amounts of data. An LLM can “understand” your questions and instructions. On its own, however, the LLM is limited to the data it was trained on. It can’t even search the web for current movie showtimes without some “tool” allowing it to perform a web search.
From the very early days of the GenAI buzz, developers have been building and adding “tools” into AI applications. Initially, the creation of these tools was ad hoc and varied depending on the developer, LLM, programming language, and the tool’s goal. But recently, a new framework for building AI tools has gotten a lot of excitement and is starting to become a new “standard” for tool development.
This framework is known as the Model Context Protocol (MCP). Originally developed by Anthropic, the company behind Claude, any developer to use MCP to build tools, called “MCP Servers,” and any AI platform can act as an “MCP Client” to use these tools. It’s essential to remember that we are still in the very early days of AI and AgenticAI; however, currently, MCP appears to be the approach for tool building. So I figured I’d dig in and figure out how MCP works by building my own very basic NetAI Agent.
I’m far from the first networking engineer to want to dive into this space, so I started by reading a couple of very helpful blog posts by my buddy Kareem Iskander, Head of Technical Advocacy in Learn with Cisco.
These gave me a jumpstart on the key topics, and Kareem was helpful enough to provide some example code for creating an MCP server. I was ready to explore more on my own.
Creating a local NetAI playground lab
There is no shortage of AI tools and platforms today. There is ChatGPT, Claude, Mistral, Gemini, and so many more. Indeed, I utilize many of them regularly for various AI tasks. However, for experimenting with agentic AI and AI tools, I wanted something that was 100% local and didn’t rely on a cloud-connected service.
A primary reason for this desire was that I wanted to ensure all of my AI interactions remained entirely on my computer and within my network. I knew I would be experimenting in an entirely new area of development. I was also going to send data about “my network” to the LLM for processing. And while I’ll be using non-production lab systems for all the testing, I still didn’t like the idea of leveraging cloud-based AI systems. I would feel freer to learn and make mistakes if I knew the risk was low. Yes, low… Nothing is completely risk-free.
Luckily, this wasn’t the first time I considered local LLM work, and I had a couple of possible options ready to go. The first is Ollama, a powerful open-source engine for running LLMs locally, or at least on your own server. The second is LMStudio, and while not itself open source, it has an open source foundation, and it is free to use for both personal and “at work” experimentation with AI models. When I read a recent blog by LMStudio about MCP support now being included, I decided to give it a try for my experimentation.
Creating Mr Packets with LMStudio
LMStudio is a client for running LLMs, but it isn’t an LLM itself. It provides access to a large number of LLMs available for download and running. With so many LLM options available, it can be overwhelming when you get started. The key things for this blog post and demonstration are that you need a model that has been trained for “tool use.” Not all models are. And furthermore, not all “tool-using” models actually work with tools. For this demonstration, I’m using the google/gemma-2-9b model. It’s an “open model” built using the same research and tooling behind Gemini.
The next thing I needed for my experimentation was an initial idea for a tool to build. After some thought, I decided a good “hello world” for my new NetAI project would be a way for AI to send and process “show commands” from a network device. I chose pyATS to be my NetDevOps library of choice for this project. In addition to being a library that I’m very familiar with, it has the benefit of automatic output processing into JSON through the library of parsers included in pyATS. I could also, within just a couple of minutes, generate a basic Python function to send a show command to a network device and return the output as a starting point.
Between Kareem’s blog posts and the getting-started guide for FastMCP 2.0, I learned it was frighteningly easy to convert my function into an MCP Server/Tool. I just needed to add five lines of code.
from fastmcp import FastMCP
mcp = FastMCP("NetAI Hello World")
@mcp.tool()
def send_show_command()
.
.
if __name__ == "__main__":
mcp.run()
Well.. it was ALMOST that easy. I did have to make a few adjustments to the above basics to get it to run successfully. You can see the full working copy of the code in my newly created NetAI-Learning project on GitHub.
As for those few adjustments, the changes I made were:
A nice, detailed docstring for the function behind the tool. MCP clients use the details from the docstring to understand how and why to use the tool.
After some experimentation, I opted to use “http” transport for the MCP server rather than the default and more common “STDIO.” The reason I went this way was to prepare for the next phase of my experimentation, when my pyATS MCP server would likely run within the network lab environment itself, rather than on my laptop. STDIO requires the MCP Client and Server to run on the same host system.
So I fired up the MCP Server, hoping that there wouldn’t be any errors. (Okay, to be honest, it took a couple of iterations in development to get it working without errors… but I’m doing this blog post “cooking show style,” where the boring work along the way is hidden. 😉
The next step was to configure LMStudio to act as the MCP Client and connect to the server to have access to the new “send_show_command” tool. While not “standardized, “most MCP Clients use a very common JSON configuration to define the servers. LMStudio is one of these clients.
Adding the pyATS MCP server to LMStudio
Wait… if you’re wondering, ‘Where’s the network, Hank? What device are you sending the ‘show commands’ to?’ No worries, my inquisitive friend: I created a very simple Cisco Modeling Labs (CML) topology with a couple of IOL devices configured for direct SSH access using the PATty feature.
NetAI Hello World CML Network
Let’s see it in action!
Okay, I’m sure you are ready to see it in action. I know I sure was as I was building it. So let’s do it!
To start, I instructed the LLM on how to connect to my network devices in the initial message.
Telling the LLM about my devices
I did this because the pyATS tool needs the address and credential information for the devices. In the future I’d like to look at the MCP servers for different source of truth options like NetBox and Vault so it can “look them up” as needed. But for now, we’ll start simple.
First question: Let’s ask about software version info.
You can see the details of the tool call by diving into the input/output screen.
This is pretty cool, but what exactly is happening here? Let’s walk through the steps involved.
The LLM client starts and queries the configured MCP servers to discover the tools available.
I send a “prompt” to the LLM to consider.
The LLM processes my prompts. It “considers” the different tools available and if they might be relevant as part of building a response to the prompt.
The LLM determines that the “send_show_command” tool is relevant to the prompt and builds a proper payload to call the tool.
The LLM invokes the tool with the proper arguments from the prompt.
The MCP server processes the called request from the LLM and returns the result.
The LLM takes the returned results, along with the original prompt/question as the new input to use to generate the response.
The LLM generates and returns a response to the query.
This isn’t all that different from what you might do if you were asked the same question.
You would consider the question, “What software version is router01 running?”
You’d think about the different ways you could get the information needed to answer the question. Your “tools,” so to speak.
You’d decide on a tool and use it to gather the information you needed. Probably SSH to the router and run “show version.”
You’d review the returned output from the command.
You’d then reply to whoever asked you the question with the proper answer.
Hopefully, this helps demystify a little about how these “AI Agents” work under the hood.
How about one more example? Perhaps something a bit more complex than simply “show version.” Let’s see if the NetAI agent can help identify which switch port the host is connected to by describing the basic process involved.
Here’s the question—sorry, prompt, that I submit to the LLM:
Prompt asking a multi-step question of the LLM.
What we should notice about this prompt is that it will require the LLM to send and process show commands from two different network devices. Just like with the first example, I do NOT tell the LLM which command to run. I only ask for the information I need. There isn’t a “tool” that knows the IOS commands. That knowledge is part of the LLM’s training data.
Let’s see how it does with this prompt:
The LLM successfully executes the multi-step plan.
And look at that, it was able to handle the multi-step procedure to answer my question. The LLM even explained what commands it was going to run, and how it was going to use the output. And if you scroll back up to the CML network diagram, you’ll see that it correctly identifies interface Ethernet0/2 as the switch port to which the host was connected.
So what’s next, Hank?
Hopefully, you found this exploration of agentic AI tool creation and experimentation as interesting as I have. And maybe you’re starting to see the possibilities for your own daily use. If you’d like to try some of this out on your own, you can find everything you need on my netai-learning GitHub project.
The mcp-pyats code for the MCP Server. You’ll find both the simple “hello world” example and a more developed work-in-progress tool that I’m adding additional features to. Feel free to use either.
The CML topology I used for this blog post. Though any network that is SSH reachable will work.
A “System Prompt Library” where I’ve included the System Prompts for both a basic “Mr. Packets” network assistant and the agentic AI tool. These aren’t required for experimenting with NetAI use cases, but System Prompts can be useful to ensure the results you’re after with LLM.
A couple of “gotchas” I wanted to share that I encountered during this learning process, which I hope might save you some time:
First, not all LLMs that claim to be “trained for tool use” will work with MCP servers and tools. Or at least the ones I’ve been building and testing. Specifically, I struggled with Llama 3.1 and Phi 4. Both seemed to indicate they were “tool users,” but they failed to call my tools. At first, I thought this was due to my code, but once I switched to Gemma 2, they worked immediately. (I also tested with Qwen3 and had good results.)
Second, once you add the MCP Server to LMStudio’s “mcp.json” configuration file, LMStudio initiates a connection and maintains an active session. This means that if you stop and restart the MCP server code, the session is broken, giving you an error in LMStudio on your next prompt submission. To fix this issue, you’ll need to either close and restart LMStudio or edit the “mcp.json” file to delete the server, save it, and then re-add it. (There is a bug filed with LMStudio on this problem. Hopefully, they’ll fix it in an upcoming release, but for now, it does make development a bit annoying.)
As for me, I’ll continue exploring the concept of NetAI and how AI agents and tools can make our lives as network engineers more productive. I’ll be back here with my next blog once I have something new and interesting to share.
In the meantime, how are you experimenting with agentic AI? Are you excited about the potential? Any suggestions for an LLM that works well with network engineering knowledge? Let me know in the comments below. Talk to you all soon!
MANILA, Philippines — Long-time transport leader George San Mateo passed away, the Pagkakaisa ng mga Samahan ng Tsuper at Opereytor Nationwide (Piston) announced on Saturday. He was 60.
Piston said San Mateo, who served as its former national president, died due to a heart attack around 9 p.m. on Friday.
“Ka George was known across the country as a fearless leader and activist fighting for the rights and welfare of jeepney drivers, operators, and working people,” Piston said in a statement.
Article continues after this advertisement
According to the transport group, San Mateo was born and raised in Metro Manila.
“As a teenager during the Marcos Sr. dictatorship, Ka George witnessed injustice early on,” Piston said, noting that this led him to join Kabataan para sa Demokrasya at Nasyonalismo (Kadena), which is “a youth group that stood up for democracy and national sovereignty during the dying years of the Marcos Sr. regime.”
In 1985, San Mateo served as the chair of the Kadena-Parañaque chapter, and in 1987, he became the group’s national spokesperson.
San Mateo worked as a driver for over a decade before joining the Piston in 2004 as a public information officer.
Article continues after this advertisement
He was later elected as Piston-Metro Manila secretary general and then appointed as national spokesperson in 2005. He became the group’s national secretary general in 2007.
In 2012, San Mateo was elected as Piston’s national president.
Article continues after this advertisement
“As president, Ka George led high-profile strikes and protests against fuel price hikes during the Aquino administration,” Piston recalled.
“Under his stewardship, Piston became known for militant yet publicly resonant campaigns that challenge both corporate and foreign interests and government inaction on public transport issues,” it added.
In 2017, San Mateo also led the transport group’s fight against the Duterte administration’s Public Utility Vehicle Modernization Program, “pushing instead for a just, people-centered public transport policy rather than a profit-driven phaseout of traditional jeepneys.”
He was arrested for leading a two-day nationwide transport strike in December 2017 and was only released on bail and eventually cleared of charges in 2020.
“These legal actions underscored the Philippine state’s contentious relationship with transport workers and activists resisting anti-people policies,” said Piston.
Your subscription could not be saved. Please try again.
Your subscription has been successful.
“After serving as Piston’s national president, Ka George was succeeded by Mody Floranda in 2019 and has since held the title of president emeritus,” it added. /cb
Recovery from surgery is a time when comfort, rest, and emotional support matter most. While flowers and greeting cards are traditional options, there are more thoughtful gifts that can genuinely assist with healing and improve the patient’s daily experience. Choosing items that promote physical relief, reduce stress, or help pass the time can make a meaningful difference.
Comfort-First Essentials
After surgery, the body needs time to heal, and rest is essential. Gifts that improve comfort, such as a soft neck pillow, an adjustable lap desk, or a warm blanket, help the patient stay relaxed while resting in bed or on the couch. Slip-resistant socks, supportive pillows, and lightweight robes are also practical items that can ease day-to-day movement and recovery.
For surgeries involving limited mobility, consider items that help with independence. Reaching tools, water bottles with straws, or pre-packaged healthy snacks make it easier for the patient to manage without constant assistance.
Entertainment And Distraction
Long recovery periods can lead to boredom and frustration. Books, audiobooks, or subscription services for streaming movies and shows provide distraction and entertainment. Adult coloring books or puzzle books are good low-effort activities that can help pass the time while still keeping the mind engaged.
Journals are another thoughtful gift, allowing patients to track their recovery progress, express emotions, or simply document daily experiences. Writing can support both mental health and a sense of control during a period that often feels uncertain.
Soothing Food Gifts
If dietary restrictions allow, small food gifts can lift spirits and encourage better nutrition. Comforting herbal teas, fruit baskets, or homemade soups are helpful options that show care. Seasonal items like fall caramel apples offer a gentle treat and bring a sense of normalcy, especially during autumn recoveries.
Personalized Support
Sometimes, the most helpful gifts are those that meet specific needs. Offering to help with errands, child care, or pet care can be more valuable than any item. A written schedule for medications or physical therapy appointments can also support recovery in a practical way.
Thoughtful gifts during post-surgery healing is less about cost and more about consideration. Items that support rest, ease discomfort, or offer a moment of joy can have a lasting impact on recovery. For more information, look over the infographic below.
Ever wonder why everything is sold as a monthly payment? It’s not an accident.
Marketers have realized that if you take a big price and break it down into a series of smaller, more palatable payments, we are more likely to buy something. It’s called the Monthly Money Trap.
That’s why a real estate agent will say that after a 10% down payment and a 30 year loan of 6.5%, the monthly payment is less than $3,000. (assumes $3,000 in property taxes and an annual $1,500 home insurance premium)
A $416,900 home with a 30 year mortgage? That’s scary.
But a $3,000 monthly payment? That’s doable. And that’s the trick.
But it’s also where the trap comes in.
Reframing total cost makes expensive things feel affordable.
It’s called the monthly money trap.
The Psychology Behind Monthly Payments
The monthly money trap is how we break down total cost into a monthly payment and then convince ourselves we can afford it. Or someone else convinces us we can afford it.
This is how the trap works. The human brain is bad at long term planning. We can imagine how life will be in a week. It’s pretty good at imagining what it’ll be like in a year. But extrapolate it out beyond that and it’s hard.
What will life be like in five years? Ten? If you had asked 20-year-old Jim what life would be like at 30, he would’ve gotten it wrong. At 40? Forget it.
Salespeople understand this. So you take a very big purchase, break it down into easy to digest monthly payments, and you can better understand how it fits in your budget.
In reality, we should look at the total cost of ownership and assess what that does to our finances.
On its own, this is not bad. This breakdown can help with planning, but only if you zoom out.
But you don’t stop with the monthly cost and make a decision based on that.
If you do, you can be convinced to spend more in total because the monthly payment is OK. You can play with the purchase, adjusting different factors, but the monthly cost only goes up a little bit.
How Car Dealerships Use This Trap
Car dealerships are famous for this. Ignore the sticker price, ignore the total cost of ownership, ignore the fuel efficiency, and just look at the monthly payment.
In fact, they will play games with all the different loan terms to get to a monthly payment you will accept. They adjust the length of the loan, the interest rate, the amount of your trade-in or down-payment, and even throw in incentives… all to get you to say yes.
If you can afford to pay $750 a month on a car, here’s how the loan term affects the price you can pay with a 5% APR loan:
As you can see, you can afford more vehicle the longer the loan, but you pay more in interest as well.
Also, remember that’s just the sticker price. This doesn’t consider other costs like insurance, fuel, routine maintenance, etc. For that, Kelley Blue Book and other resources are good for figuring that out for your target vehicle.
How Do You Avoid This?
You must recognize the tactic when someone uses it on you. Just like how you need to recognize someone trying to use scare tactics and scarcity (time is running out! It’s the last one! etc.), the monthly trap is a tactic too.
Always look at the total cost first. With the car example above, we can see that all three loan terms were supported by a $750 monthly payment.
The question you need to ask yourself is whether you want to pay all that interest to get into a higher priced car. If your plan is to switch cars every five years, getting a five year loan may not be the best idea for you. By the time you’ve paid off the loan, the value of the car will have fallen very far from $39,750.
KBB says new cars depreciate 30% over the first two years and then 8-12% each year after that. Assuming it only depreciates 8% a year after the first two years, your $39,750 car is worth only $21,667 – a loss of value of $18,083.
If you plan on driving the car into the ground, which could take 15 years, then depreciation isn’t an issue. The $48,385 spread across 15 years which makes it a mere $3,225 a year or $268 a month. Even when you add in the other variable costs (insurance, fuel, etc.), it still makes sense.
So the next time someone tries to sell you on a purchase with the monthly cost, you’re prepared.
Your monthly payment is just one piece of the puzzle. Before you commit, ask yourself what the purchase really costs and whether it’s something you want in your future plans.
