Minimizing your Data Attack Surface in the Cloud

The LLM Hype, and What’s Missing

Over the past few years, large language models (LLMs) have dominated the AI conversation. From writing essays to generating code, LLMs like GPT-4 and Claude have proven that massive models can produce human-like language and reasoning at scale.

‍

But here's the catch: not every task needs a 70-billion-parameter model. Parameters are computationally expensive - they require both memory and processing time.

‍

At Sentra, we discovered early on that the work our customers rely on for accurate, scalable classification of massive data flows - isn’t about writing essays or generating text. It’s about making decisions fast, reliably, and cost-effectively across dynamic, real-world data environments. While large language models (LLMs) are excellent at solving general problems, it creates a lot of unnecessary computational overhead.

‍

That’s why we’ve shifted our focus toward Small Language Models (SLMs) - compact, specialized models purpose-built for a single task - understanding and classifying data efficiently. By running hundreds of SLMs in parallel on regular CPUs, Sentra can deliver faster insights, stronger data privacy, and a dramatically lower total cost of AI-based classification that scales with their business, not their cloud bill.

What Is an SLM?

An SLM is a smaller, domain-specific version of a language model. Instead of trying to understand and generate any kind of text, an SLM is trained to excel at a particular task, such as identifying the topic of a document (what the document is about or what type of document it is), or detecting sensitive entities within documents, such as passwords, social security numbers, or other forms of PII.

‍

In other words: If an LLM is a generalist, an SLM is a specialist. At Sentra, we use SLMs that are tuned and optimized for security data classification, allowing them to process high volumes of content with remarkable speed, consistency, and precision. These SLMs are based on standard open source models, but trained with data that was curated by Sentra, to achieve the level of accuracy that only Sentra can guarantee.

From LLMs to SLMs: A Strategic Evolution

Like many in the industry, we started by testing LLMs to see how well they could classify and label data. They were powerful, but also slow, expensive, and difficult to scale. Over time, it became clear: LLMs are too big and too expensive to run on customer data for Sentra to be a viable, cost effective solution for data classification.

Each SLM handles a focused part of the process: initial categorization, text extraction from documents and images, and sensitive entity classification. The SLMs are not only accurate (even more accurate than LLMs classifying using prompts) - they can run on standard CPUs efficiently, and they run inside the customer’s environment, as part of Sentra’s scanners.

The Benefits of SLMs for Customers

a. Speed and Efficiency

SLMs process data faster because they’re lean by design. They don’t waste cycles generating full sentences or reasoning across irrelevant contexts. This means real-time or near-real-time classification, even across millions of data points.

b. Accuracy and Adaptability

SLMs are pre-trained “zero-shot” language models that can categorize and classify generically, without the need to pre-train on a specific task in advance. This is the meaning of “zero shot” - it means that regardless of the data it was trained on, the model can classify an arbitrary set of entities and document labels without training on each one specifically. This is possible due to the fact that language models are very advanced, and they are able to capture deep natural language understanding at the training stage.

‍

Regardless of that, Sentra fine tunes these models to further increase the accuracy of the classification, by curating a very large set of tagged data that resembles the type of data that our customers usually run into.

Our feedback loops ensure that model performance only gets better over time - a direct reflection of our customers’ evolving environments.

c. Cost and Sustainability

Because SLMs are compact, they require less compute power, which means lower operational costs and a smaller carbon footprint. This efficiency allows us to deliver powerful AI capabilities to customers without passing on the heavy infrastructure costs of running massive models.

d. Security and Control

Unlike LLMs hosted on external APIs, SLMs can be run within Sentra’s secure environment, preserving data privacy and regulatory compliance. Customers maintain full control over their sensitive information - a critical requirement in enterprise data security.

A Quick Comparison: SLMs vs. LLMs

The difference between SLMs and LLMs becomes clear when you look at their performance across key dimensions:

‍

‍

Continuous Learning: How Sentra’s SLMs Grow

One of the most powerful aspects of our SLM approach is continuous learning. Each Sentra customer project contributes valuable insights, from new data patterns to evolving classification needs. These learnings feed back into our training workflows, helping us refine and expand our models over time.

‍

While not every model retrains automatically, the system is built to support iterative optimization: as our team analyzes feedback and performance, models can be fine-tuned or extended to handle new categories and contexts.

The result is an adaptive ecosystem of SLMs that becomes more effective as our customer base and data diversity grow, ensuring Sentra’s AI remains aligned with real-world use cases.

Sentra’s Multi-SLM Architecture

Sentra’s scanning technology doesn’t rely on a single model. We run many SLMs in parallel, each specializing in a distinct layer of classification:

‍

1. Embedding models that convert data into meaningful vector representations
2. Entity Classification models that label sensitive entities
3. Document Classification models that label documents by type
4. Image-to-text and speech-to-text models that are able to process non-textual data into textual data

This layered approach allows us to operate at scale - quickly, cheaply, and with great results. In practice, that means faster insights, fewer errors, and a more responsive platform for every customer.

The Future of AI Is Specialized

We believe the next frontier of AI isn’t about who can build the biggest model, it’s about who can build the most efficient, adaptive, and secure ones.

‍

By embracing SLMs, Sentra is pioneering a future where AI systems are purpose-built, transparent, and sustainable. Our approach aligns with a broader industry shift toward task-optimized intelligence - models that do one thing extremely well and can learn continuously over time.

Conclusion: The Power of Small

At Sentra, we’ve learned that in AI, bigger isn’t always better. Our commitment to SLMs reflects our belief that efficiency, adaptability, and precision matter most for customers. By running thousands of small, smart models rather than a single massive one, we’re able to classify data faster, cheaper, and with greater accuracy - all while ensuring customer privacy and control.

‍

In short: Sentra’s SLMs represent the power of small, and the future of intelligent classification.

<blogcta-big>

Finding My Voice: From Kenya to the Global Stage

I was born and raised in Kenya, the youngest of three and the only daughter. My parents, who never had the chance to finish their education, sacrificed everything to give me opportunities they never had. Their courage became my foundation.

‍

At sixteen, my mother signed me up to speak at a community event, without telling me first! I stood before 500 people and spoke about something that had long bothered me: there were no women on our community board. That same year, two women were appointed for the first time in our community’s history. This year, I was given the recognition for being a Community Leader at the Global Gujrati Gaurav Awards in BC for my work in educating seniors on cyber safety and helping many immigrants secure jobs.

‍

I didn’t realize it then, but that moment would define my purpose: to speak up for those whose voices aren’t always heard.

From Isolation to Empowerment

When I moved to the UK to study Financial Economics, I faced a different kind of challenge - isolation. My accent made me stand out, and not always in a good way. There were times I felt invisible, even rejected. But I made a promise to myself in those lonely moments that no one else should feel the same way.

‍

Years later, as a founding member of WiCyS Western Affiliate, I helped redesign how networking happens at cybersecurity events. Instead of leaving it to chance, we introduced structured networking that ensured everyone left with at least one new connection. It was a small change, but it made a big difference. Today, that format has been adopted by organizations like ISC2 and ISACA, creating spaces where every person feels they belong. 

Breaking Barriers and Building SHE

When I pivoted into cybersecurity sales after moving to Canada, I encountered another wall. I applied for a senior role and failed a personality test, one that unfairly filtered out many talented women. I refused to accept that. I focused on listening, solving real customer challenges, and eventually became the top seller. That success helped eliminate the test altogether, opening doors for many more women who came after me. That experience planted a seed that would grow into one of my proudest initiatives: SHE (Sharing Her Empowerment).

‍

It started as a simple fireside chat on diversity and inclusion - just 40 seats over lunch. Within minutes of sending the invite, we had 90 people signed up. Executives moved us into a larger room, and that event changed everything. SHE became our first employee resource group focused on empowering women, increasing representation in leadership, and amplifying women’s voices within the organization. Even with just 19% women, we created a ripple effect that reached the boardroom and beyond.

‍

SHE showed me that when women stand together, transformation happens.

Creating Pathways for the Next Generation

Mentorship has always been close to my heart. During the pandemic, I met incredible women, who were trying to break into cybersecurity but kept facing barriers. I challenged hiring norms, advocated for fair opportunities, and helped launch internship programs that gave women hands-on experience. Today, many of them are thriving in their cyber careers, a true reflection of what’s possible when we lift as we climb.

‍

Through Standout to Lead, I partnered with Women Get On Board to help women in cybersecurity gain board seats. Watching more women step into decision-making roles reminds me that leadership isn’t about titles, it’s about creating pathways for others.

Women in Cybersecurity: Our Collective Story

This year, I’m deeply honored to be named among the Top 20 Cybersecurity Women of the World by the United Cybersecurity Alliance. Their mission - to empower women, elevate diverse voices, and drive equity in our field, mirrors everything I believe in.

‍

I’m also thrilled to be part of the upcoming documentary premiere, “The WOMEN IN SECURITY Documentary,” proudly sponsored by Sentra, Amazon WWOS, and Pinkerton among others. This film shines a light on the fearless women redefining what leadership looks like in our industry.

‍

As a member of Sentra’s community, I see the same commitment to visibility, inclusion, and impact that has guided my journey. Together, we’re not just securing data, we’re securing the future of those who will lead next.

Asante Sana – Thank You

My story, my safari, is still being written. I’ve learned that impact doesn’t come from perfection, but from purpose. Whether it’s advocating for fairness, mentoring the next generation, or sharing our stories, every step we take matters.

To every woman, every underrepresented voice in STEM, and everyone who’s ever felt unseen - stay authentic, speak up, and don’t be afraid of the outcome. You might just change the world.

‍

Join me and the Sentra team at The WOMEN IN SECURITY Documentary Premiere, a celebration of leadership, resilience, and the voices shaping the future of our industry.

‍

Save your seat at The Women in Security premiere here (spots are limited).

‍

Follow Sentra on LinkedIn and YouTube for more updates on the event and stories that inspire change.

<blogcta-big>

Why Data Visibility Belongs in Your 2026 Cybersecurity Budget

As the fiscal year winds down and security leaders tackle cybersecurity budget planning for 2026, you need to decide how to use every remaining 2025 dollar wisely and how to plan smarter for next year. The question isn’t just what to cut or keep, it’s what creates measurable impact. Across programs, data visibility and DSPM deliver provable risk reduction, faster audits, and clearer ROI,making them priority line items whether you’re spending down this year or shaping next year’s plan. Some teams discover unspent funds after project delays, postponed renewals, or slower-than-expected hiring. Others are already deep in planning mode, mapping next year’s security priorities across people, tools, and processes. Either way, one question looms large: where can a limited security budget make the biggest impact - right now and next year?

‍

Across the industry, one theme is clear: data visibility is no longer a “nice-to-have” line item, it’s a foundational control. Whether you’re allocating leftover funds before year-end or shaping your 2026 strategy, investing in Data Security Posture Management (DSPM) should be part of the plan.

‍

As Bitsight notes, many organizations look for smart ways to use remaining funds that don’t roll over. The goal isn’t simply to spend, it’s to invest in initiatives that improve posture and provide measurable, lasting value. And according to Applied Tech, “using remaining IT funds strategically can strengthen your position for the next budget cycle.”

‍

That same principle applies in cybersecurity. Whether you’re closing out this year or planning for 2026, the focus should be on spending that improves security maturity and tells a story leadership understands. Few areas achieve that more effectively than data-centric visibility.

‍

(For additional background, see Sentra’s article on why DSPM should take a slice of your cybersecurity budget.)

Where to Allocate Remaining Year-End Funds (Without Hurting Next Year’s Budget)

It’s important to utilize all of your 2025 budget allocations because finance departments frequently view underspending as a sign of overfunding, leading to smaller allocations next year. Instead, strategic security teams look for ways to convert every remaining dollar into evidence of progress.

‍

That means focusing on investments that:

• Produce measurable results you can show to leadership.
  
• Strengthen core program foundations: people, visibility, and process.
  
• Avoid new recurring costs that stretch future budgets.
  

Top Investments That Pay Off

1. Invest in Your People

One of the strongest points echoed by security professionals across industry communities: the best investment is almost always your people. Security programs are built on human capability. Certifications, practical training, and professional growth not only expand your team’s skills but also build morale and retention, two things that can’t be bought with tooling alone.

‍

High-impact options include:

‍

• Hands-on training platforms like Hack The Box, INE Skill Dive, or Security Blue Team, which develop real-world skills through simulated environments.
  
• Professional certifications (SANS GIAC, OSCP, or cloud security credentials) that validate expertise and strengthen your team’s credibility.
  
• Conference attendance for exposure to new threat perspectives and networking with peers.
  
• Cross-functional training between SOC, GRC, and AppSec to create operational cohesion.
  
  

In practitioner discussions, one common sentiment stood out: training isn’t just an expense, it’s proof of leadership maturity.

As one manager put it, “If you want your analysts to go the extra mile during an incident, show you’ll go the extra mile for them when things are calm.”

2. Invest in Data Visibility (DSPM)

While team capability drives execution, data visibility drives confidence. In recent conversations among mid-market and enterprise security teams, Data Security Posture Management (DSPM) repeatedly surfaced as one of the most valuable investments made in the past year, especially for hybrid-cloud environments.

‍

One security leader described it this way:

“After implementing DSPM, we finally had a clear picture of where sensitive data actually lived. It saved our team hours of manual chasing and made the audit season much easier.”

‍

That feedback reflects a growing consensus: without visibility into where sensitive data resides, who can access it, and how it’s secured, every other layer of defense operates partly in the dark.

*Tip: If your remaining 2025 budget won’t suffice for a full DSPM deployment, you can scope an initial implementation with the remaining budget, then expand to full coverage in 2026.

‍

DSPM solutions provide that clarity by helping teams:

‍

• Map and classify sensitive data across multi-cloud and SaaS environments.
  
• Identify access misconfigurations or risky sharing patterns.
  
• Detect policy violations or overexposure before they become incidents.
  
  

Beyond security operations, DSPM delivers something finance and leadership appreciate, measurable proof. Dashboards and reports make risk tangible, allowing CISOs to demonstrate progress in data protection and compliance.

‍

The takeaway: DSPM isn’t just a good way to use remaining funds, it’s a baseline investment every forward-looking security program should plan for in 2026 and beyond.

3. Invest in Testing

‍

Training builds capability. Visibility builds understanding. Testing builds credibility.

External red team, purple team, or security posture assessments continue to be among the most effective ways to validate your defenses and generate actionable findings.

‍

Security practitioners often point out that testing engagements create outcomes leadership understands:

‍

“Training is great, but it’s hard to quantify. An external assessment gives you findings, metrics, and a roadmap you can point to when defending next year’s budget.”

‍

Well-scoped assessments do more than uncover vulnerabilities—they benchmark performance, expose process gaps, and generate data-backed justification for continued investment.

4. Preserve Flexibility with a Retainer

If your team can’t launch a new project before year-end, a retainer with a trusted partner is an efficient way to preserve funds without waste. Retainers can cover services like penetration testing, incident response, or advisory hours, providing flexibility when unpredictable needs arise. This approach, often recommended by veteran CISOs, allows teams to close their books responsibly while keeping agility for the next fiscal year.

5. Strengthen Your Foundations

Not every valuable investment requires new tools. Several practitioners emphasized the long-term returns from process improvements and collaboration-focused initiatives:

‍

• Threat modeling workshops that align development and security priorities.
  
• Framework assessments (like NIST CSF or ISO 27001) that provide measurable baselines.
  
• Automation pilots to eliminate repetitive manual work.
  
• Internal tabletop exercises that enhance cross-team coordination.
  
  

These lower-cost efforts improve resilience and efficiency, two metrics that always matter in budget conversations.

How to Decide: A Simple, Measurable Framework

When evaluating where to allocate remaining or future funds, apply a simple framework:

‍

1. Identify what’s lagging. Which pillar - people, visibility, or process most limits your current effectiveness?
   
2. Choose something measurable. Prioritize initiatives that produce clear, demonstrable outputs: reports, dashboards, certifications.
   
3. Aim for dual impact. Every investment should strengthen both your operations and your ability to justify next year’s funding.
   
   

Final Thoughts

A strong security budget isn’t just about defense, it’s about direction. Every spend tells a story about how your organization prioritizes resilience, efficiency, and visibility.

‍

Whether you’re closing out this year’s funds or preparing your 2026 plan, focus on investments that create both operational value and executive clarity. Because while technologies evolve and threats shift, understanding where your data is, who can access it, and how it’s protected remains the cornerstone of a mature security program.

‍

Or, as one practitioner summed it up: “Spend on the things that make next year’s budget conversation easier.”

DSPM fits that description perfectly.

<blogcta-big>