Jatheon Technologies Inc.

Jatheon empowers 500+ organizations to store, access, retrieve and manage email, social media and mobile communications for compliance and eDiscovery.

Jatheon Cloud is perfectly suited for small to medium-sized businesses offering unparalleled flexibility and bottomless storage that solves all your archiving, compliance and eDiscovery challenges at once.

Jatheon appliances include powerful on-premise email archiving and governance solutions  with advanced search and retrieval software, specialized hardware, 24/7 remote monitoring and technical support and free-of-charge hardware replacement every 4 years to ensure the optimum operation of your business.

Started as a Sr. Software Engineer and held various roles within the organization including Professional Services Manager, Head of Technical Support, VP Sales and Technical Support, VP Digital Marketing and VP Operations.

Various Roles

Conics Inc.

Principal

Cygnal Technologies Inc.

Manager, Software Solutions

Edaptivity inc.

Acquired by Cygnal Technologies in 2003 for $2.5 million.

Software Engineer

Hey! Do you know how important it is today for companies to archive everything from emails to social posts? 

Each year, new laws and regulations pop up about data archiving. This makes it harder for companies to keep up.

What you need is an archiving solution to help you store your communications data and be able to find it quickly (all of us know how short legal deadlines can be).

I’m the CEO of Jatheon Technologies, and I’m here to show you that the archiving industry doesn’t have to be boring.

Jatheon Cloud is the fourth-generation archiving solution that helps you get your data like email, social media, chat apps, and text messages captured, indexed, retained, and searchable. 

Here’s why we should be on your radar:


📧 We’ve processed 11+ billion messages 
🏆 We’re the best-rated data archiving product on Gartner (4.9 out of 5)
☁ Our product is built using AWS micro services 
🔎 You can search 1M communication records in <1 second
🏢 We’re the vendor of choice for 500+ schools, government agencies, banks, and healthcare providers

If you want to know more about the data archiving industry, ediscovery, and the lifestyle of a CEO, give me a follow (and ping the bell to receive updates once I post).

Bezos is running a company again. Didn’t see that coming.

He’s co-CEO of Project Prometheus, an AI startup that just raised $6.2 billion. First time he’s actually run anything since stepping down from Amazon four years ago. Not advising, not investing from the sidelines. Actually running it.

They’re going after AI for manufacturing and engineering. The pitch is AI that learns by building physical things and screwing up in the world, not just reading the internet and regurgitating text.

They poached 100 people from OpenAI, DeepMind, and Meta. His co-CEO is Vik Bajaj from Google X, the guy who worked on Waymo and Wing. They're not messing around with talent.

Here’s what I keep thinking about. Bezos has more money than he could spend in ten lifetimes. He could be doing literally anything. Instead, he’s back to working 80 hour weeks at a startup trying to solve one of the hardest problems in AI.

Either he’s bored or he actually believes this is the next big thing. My guess is both.

Most of these mega-funded AI companies flame out. But when someone like Bezos puts his skin in the game, not just money but time, you have to pay attention. He’s built two massive companies already. He knows what it takes.

Can’t wait to find out if they can actually deliver or if this is just another pile of cash-chasing hype.

Marko Dinic

CEO @Jatheon | Building the fastest AI-powered data archive so your company can act faster, smarter, safer

Cloudflare went down this morning. X, ChatGPT, Spotify, and even McDonald's kiosks stopped working. Error 500 everywhere.

Third major outage in a month. AWS took down government sites and banks for 15 hours a few weeks ago. Microsoft Azure before that. Now Cloudflare. Different companies, same issue. Too much of the internet depends on too few providers.

The funny part is DownDetector went down too because it runs on Cloudflare. The site you check when things break was broken.

Everyone talks about multi-cloud strategy after these outages but when everything comes back online nobody wants to spend the money….. Our nature is to optimize for cost over reliability. Until your infrastructure goes down and you're stuck explaining to customers why nothing works.

We just shipped YouTube and SharePoint archiving. Little late to the party, but we never really saw demand for it outside the financial services vertical. But now we need as many data sources as possible to allow our AI (Liya) to cross reference for meaningful and complete answers to all questions. 

Privacy junkies won’t like it, but corporate AI is basically already here and becoming more useful every day. We put a lot of effort to protect privacy in the AI age, and basically built the best and most granular Role and Permissions access around.

No alternative text description for this image

Prince Harry and 800 others just signed a letter calling for a ban on AI superintelligence. Who else signed? Geoffrey Hinton, Steve Bannon, and Steve Wozniak - quite the coalition. The statement calls for a prohibition on developing superintelligence until there is broad scientific consensus it can be done safely.

The companies actually building superintelligence didn’t sign it. Sam Altman says superintelligence will arrive by 2030, Mark Zuckerberg says it’s “now in sight,” and neither is slowing down. They’re in a race where whoever stops loses. Max Tegmark from MIT said these companies are “stuck in this race to the bottom” with irresistible pressure to keep going.

Here’s what makes this weird - these same companies are inflating what their AI can do. OpenAI recently claimed ChatGPT solved unsolved math problems when it just summarized what was already online. So we’re simultaneously being told AI is about to surpass humanity while watching companies exaggerate basic capabilities.

Meanwhile, the real AI problems are happening now. Bias in algorithms, training on stolen data, compliance chaos, and energy waste requiring nuclear plants. But the debate focuses on theoretical threats while actual harms pile up. 

This letter won’t stop anything - open letters never do. You need regulation with teeth, not polite requests from concerned scientists.

https://hubs.ly/Q03Q9K8y0

People love WhatsApp. 

It’s fast, convenient, and they’re already using it for everything else in their lives. On the other hand, regulated organizations face the pressure to meet customers where they are, which sounds reasonable until you try to make it compliant.

The problem is that end-to-end encryption makes WhatsApp private and secure, which people love, but that same encryption makes monitoring and retention nearly impossible with standard tools. We see this tension constantly at Jatheon, and the usual response is to ban WhatsApp entirely with a strict policy.

But that doesn’t work. Employees violate the policy anyway because clients prefer WhatsApp, and now you’ve got unmonitored communication happening in the shadows, which is worse than having no policy at all. Would love to see a Compliance API from Meta. 

In the meantime, we have a working solution that’s pretty good.

AI doctor story time:

Mom spends hours chatting with the AI about her kidney transplant symptoms, uploading lab reports, and asking for guidance on medications. Her take: “DeepSeek is more humane. Doctors are more like machines.”

Healthcare systems are so broken that a chatbot feels more attentive than a three-minute doctor visit after a two-day journey and hours in the waiting room. When you get five minutes with an overworked specialist who’s already typing your prescription before you finish talking, an AI that responds instantly with detailed explanations starts looking pretty good.

But here’s what’s missing from that picture: accountability. 

When DeepSeek recommends unnecessary tests in 90% of cases and prescribes inappropriate medications in over half, there’s no one to sue when things go wrong. When it suggests antibiotics for viral infections or expensive CT scans that aren’t clinically indicated, patients bear the cost and risk. The AI doesn't have a medical license to lose.

I’ve spent 20 years in data compliance watching organizations struggle with record-keeping and accountability. Healthcare is moving toward AI assistance whether we like it or not, but we’re doing it without the infrastructure to track what these systems recommend, verify their accuracy, or hold anyone responsible when they fail. 

The real scandal isn’t that this woman trusts AI more than doctors. It’s that both the healthcare system and the AI industry failed her differently, and we’re not addressing either problem.

https://hubs.la/Q03R6-Vb0

Open article: ‘DeepSeek is humane. Doctors are more like machines’: my mother’s worrying reliance on AI for health advice by hubs.la

‘DeepSeek is humane. Doctors are more like machines’: my mother’s worrying reliance on AI for health advice

More than 1,200 venture-backed unicorns are stuck in limbo right now. No IPOs, no acquisitions, just expensive valuations on paper.

Everyone wanted to be the next unicorn, but nobody planned for what happens when the party ends. The “move fast and break things” playbook worked when capital was cheap and growth was the only metric. Ship fast, fix later, worry about profitability never.

That built some companies, but it also created mountains of technical debt and businesses that can’t survive without constant fundraising. Some took a different path. They spent years building solid foundations before scaling hard. No quarterly board meetings demanding impossible growth. They grew without burning through cash or teams.

What worked for us to keep our clients for 10+ years and be recognized as industry leaders was doing mostly unsexy things daily: response times under five minutes, building code that didn’t require the original developer to fix, and creating processes that ran without constant intervention.

The companies that prioritized sustainable systems over headline growth aren’t scrambling for bridge rounds right now. The ones that raised at inflated valuations and burned through it chasing the next funding milestone are doing down rounds, cutting 30-40% of staff, and trying to rebuild infrastructure they skipped the first time while their revenue growth flatlines at 15%.

Your hiring process should be filtering out the people who actually do the work.

The resume tells you less than you think. Everyone claims they're a “detail-oriented team player with excellent communication skills.” But what matters more is what they built. If someone reduced server response time from 3 seconds to 400ms, that’s the skill. If they “leveraged synergies to optimize performance,” that’s nothing.

Job hopping isn’t automatically bad. Three jobs in two years because you kept getting better offers is fine. Three jobs in two years because nobody wanted to keep you is a problem. Ask why they left each role. Good candidates explain what they learned. Bad ones blame their old managers.

“Culture fit” gets misused constantly. It should mean “works well with our processes and communication style.” Too often it means “ reminds me of everyone else here.” If you can’t explain specifically what culture fit means beyond vague feelings, you’re probably using it wrong.

Here’s what predicts good hires: they explain complex work simply, they show you their code, work, or portfolio without you asking, and they ask specific questions about your tech stack and team structure. The ones who interview perfectly but can’t show you anything they’ve built usually can’t do the job.

Stop hiring based on how well someone performs in an interview. Hire based on what they've accomplished.

AWS went down Monday and I immediately get calls of sympathy and advice. Didn’t affect us much but it took HMRC, US & UK government sites, banks, and thousands of apps down. The outage lasted 15 hours, started in the Virginia data center, and took down services globally. UK tax authority infrastructure was running in US data centers. Not London. Virginia.

AWS controls about 30% of the global cloud market, so when it goes down, a third of the internet goes with it. This is what vendor lock-in looks like at scale.

Everyone talks about multi-cloud strategy but nobody actually does it because it’s expensive, complicated, and doesn’t show up on quarterly earnings until something breaks. Yes, when you put all your eggs in one basket, you’re gambling with critical infrastructure. 

Companies tell themselves AWS has 99.99% uptime, great SLA, reliable infrastructure - all true until it’s not. The problem isn’t AWS reliability but the concentration risk that is hard and very expensive to address. We’ve seen this with CrowdStrike, CloudFlare, now AWS again. Different vendor, same problem - too much dependency on too few providers.

Fixing it requires investment, planning, and admitting your current architecture has a single point of failure. Most companies won’t do that until after the outage causes significant pain, not before.

https://hubs.ly/Q03Q9LcP0

SaaS companies planned for 35% median growth in 2025. They hit 26% in 2024.

This gap tells you everything about where the industry is right now. Everyone’s making aggressive projections while the math underneath doesn’t support them.

The formula is simple: net burn divided by net new ARR. Below 1.5 is efficient. Above 2.0 means you’re spending two dollars to generate one dollar of recurring revenue. 

Companies under $5M ARR are sitting at 1.5x to 2.5x burn multiples. That’s expected when you’re still figuring out product-market fit. But companies above $20M ARR need to be under 1.5x. The data shows most aren’t hitting that.

The companies that built discipline early, where every dollar spent had to generate returns, aren’t panicking now. The ones that assumed cheap capital would last forever are trying to cut their way to profitability, which rarely works.

CAC payback periods are stretching from 6 months to 12-18 months. Digital ad costs are up, sales cycles are longer, and competition for talent makes everything more expensive. If you’re burning more cash to acquire customers who take longer to pay back their acquisition cost, you burn multiple climbs and your runway shrinks.

Companies between $3M and $20M ARR that built sustainable unit economics from the start aren’t scrambling for emergency funding right now. The ones operating at 2.0x burn multiples or higher are having difficult conversations with their boards.

The optimistic projections for 2025 are going to hit reality in the next few quarters. If you’re planning 35% growth but your burn multiple is above 2.0, the math doesn’t work. Either growth slows, burn drops, or you run out of runway.

https://bit.ly/4p81B5U

Everyone's arguing about when quantum computers will break encryption. 2035 or 2055. MITRE's report says somewhere in between, but honestly, the exact date doesn't matter. 

NIST already published post-quantum encryption standards in 2024. Three algorithms ready to deploy: ML-KEM for encryption, ML-DSA and SLH-DSA for digital signatures. These algorithms are designed to withstand quantum attacks and work on current hardware. You can start migrating today, not in some theoretical future.

Here's what gets me: 62% of security professionals are worried about quantum threats, but only 5% have a strategy. This isn't a knowledge problem—it's an action problem. The steps are straightforward: inventory your encrypted data, identify what needs long-term protection, test post-quantum algorithms in non-critical systems, plan your migration timeline. Whether quantum arrives in 10 years or 30 years, these steps stay the same.

We're already looking at this for our archiving systems. When you're storing data for 7+ years (sometimes indefinitely), you can't ignore threats that might materialize within that retention window. The timeline debate is a distraction. NIST gave us the standards. Vendors are implementing them. Now it's just a matter of getting started.

https://hubs.la/Q03Q9MtY0

Fed cut. Bank of Canada cut. Despite inflation above their own targets. Race to the bottom, and fast.

Companies spend millions on security tools and then let someone from marketing post customer testimonials with full names on Instagram.

taplio

Marko Dinic's Linkedin Analytics

Marko Dinic

Performance ScoreMarko Dinic's LinkedIn Performance

Marko Dinic's Best Posts (last 30 days)

Ready to grow your LinkedIn presence?

taplio

.css-6ytdur{background-image:linear-gradient(284deg, rgba(39, 43, 72, 1) 0%, rgba(33, 47, 183, 1) 100%);color:transparent;-webkit-background-clip:text;background-clip:text;}Marko Dinic's Linkedin Analytics

Marko Dinic

Performance Score.css-goub6s{font-size:var(--chakra-fontSizes-md);font-weight:500;color:var(--chakra-colors-gray-600);display:block;margin-top:var(--chakra-space-1);}Marko Dinic's LinkedIn Performance

Marko Dinic's Best Posts .css-1emoo84{font-size:var(--chakra-fontSizes-md);font-weight:400;color:var(--chakra-colors-gray-600);}(last 30 days)

Ready to grow your LinkedIn presence?

Marko Dinic's Linkedin Analytics

Performance ScoreMarko Dinic's LinkedIn Performance

Marko Dinic's Best Posts (last 30 days)