With a proven track record as a Chief Revenue Officer (CRO), SaaS growth consultant, and investor, I specialize in helping SaaS companies achieve scalable, predictable revenue growth through enterprise sales strategies.

Having served as CEO, COO, and CRO across global tech and consulting sectors—including cybersecurity, telecommunications, and retail—I bring a unique blend of leadership experience and operational expertise. I leverage data-driven insights, advanced CRM tools (e.g., Salesforce, HubSpot), and tailored sales strategies to shorten sales cycles, increase win rates, and unlock new revenue opportunities.

My passion lies in partnering with SaaS companies to:

Refine and scale go-to-market strategies for enterprise sales.
Penetrate new markets and close complex, high-value B2B deals.
Empower sales teams through hands-on coaching, real-time feedback, and custom workshops.

Collaboration is at the heart of my approach. I build cohesive, high-performing teams by fostering a culture of continuous learning, open communication, and mutual respect. Through my fractional leadership and targeted coaching, I enable companies to exceed revenue targets, secure long-term client relationships, and achieve sustainable growth.

Whether you’re a SaaS startup looking to scale or an established firm aiming to break into new territories, I can help you unlock your next phase of growth. Let’s connect and explore how I can support your success.

Joshua Behar

What do Marriott, SolarWinds, and the 2025 Salesloft/Drift compromise have in common?

It's not scale.
It's not zero-day exploits.
It's not even especially clever attackers.

It's the silence.

-Marriott: ~4 years undetected. 383M+ records exposed.
-SolarWinds: 9-14 months undetected. 18,000 downstream organizations hit.
-Salesloft/Drift (2025): Stolen OAuth tokens. No exploit needed. 

Just legitimate-looking access abused across 700+ Salesforce customers.
In every case, the kill chain wasn't broken. It was a highway.
Initial access --> Credential abuse -->Lateral movement --> Privilege escalation --> Exfiltration.

Step by step. Unchallenged. Uninterrupted.

The uncomfortable truth: most of these steps don't require sophisticated hacking. They require valid credentials, excessive permissions, and an architecture built to detect threats rather than prevent them.

EDR, SIEM, CNAPP, SOAR. These tools are excellent at what they do. What they do is alert you after the attacker already has a foothold.

You're not stopping the fire. You're measuring how fast it spreads.

The harder shift is architectural:
-Can privilege escalation be blocked at the policy layer before it's attempted?
-Can lateral movement be made structurally non-viable, not just monitored?
-Can Impact surface be bounded mathematically before a single misconfiguration reaches production?

Every breach is a sequence. Misconfigurations, overprivileged identities, implicit trust, flat networks. 
Each step only works because the previous one was allowed.

Enforce the architecture. Break the sequence.

That is what we built Aryon to do. Agentless cloud security guardrails that enforce prevention at the infrastructure layer before deployment. No IaC rewrites, no engineering friction, consistent protection whether your team uses Terraform, Pulumi, CLI, or makes a live console change.

We do not watch the chain. We make the next step impossible.

One question for the CISOs and cloud security leaders here:
What is the one area you are doubling down on in 2026 to shift toward prevention-first?

Curious to hear what is actually working on the ground.

#CloudSecurity #CISO #ZeroTrust #CyberSecurity #CloudArchitecture #DevSecOps
Ron Arbel Shira Zeltner Weiss

The biggest UI trend of 2026?

No UI.

Not because interfaces are disappearing.

But because the new user of the internet doesn’t need them.
For 30 years we built software for humans:
• clicking buttons
 • reading dashboards
 • following onboarding flows

Everything was optimized for eyes, thumbs, and attention spans.
But a new user just showed up.
-It doesn't use a mouse.
-It doesn't read instructions.
-It executes thousands of tasks per second.

AI agents.

And they’re trying to operate inside software designed entirely for humans.
Which creates a strange reality:
For agents, the interface isn’t the product.
It’s friction.
Every login screen, CAPTCHA, and loading spinner is just latency.
Most teams are asking:
“How do we redesign our UI for AI?”
Wrong question.
The real one is:
“Can an agent actually operate inside our product?”
If the answer is no, you don’t have a design problem.
You have an architecture problem.

For two decades SaaS optimized UX — User Experience.
The next wave will optimize TX — Transaction Experience.

How easily can autonomous systems actually get work done?
That’s the layer we’re building at Anchor Browser.
Because the internet was built for humans.
The next layer will let agents operate inside it.
#AI #SaaS #SoftwareArchitecture #AgentNative #FutureOfWork
Idan Raman ⚓️ Dor Dankner ⚓️Guy Ben Simhon ⚓

Personize.ai

We make Gen AI adoption simple. With Personize, companies can deploy custom AI agents in hours without costly engineering.

At Personize.ai, we remove barriers to Generative AI. Our customers have produced 200 million words of high-quality, error-free content pieces (e.g. email, LinkedIn messages, Landing Pages, Blog Posts, Report Sections, Analyses, etc.).

You won't find world-class agents like ours readily available in the market. Others may keep their top agents hidden, but we're different – we build and make them available to businesses. This helps companies save time, and achieve the highest quality, accuracy, and customization in the market.

Personize.ai is a no-code platform enabling autonomous, multi-agent workflows that process thousands of records.

Our AI agents:
- Read and fetch data
- Reason and make decisions
- Write and sync results

With our "Agents as APIs" (alpha), you can integrate autonomous agents into your apps with minimal code.

Founder & Product Lead

Robust Choice (Personize AI)

We bring data science and the most cutting-edge AI capabilities to B2B sales. We are an experienced team of data scientists, product builders, and serial entrepreneurs on a mission to utilize AI and technology for sales teams.

Our product is Personize AI which is the first AI personalization engine to empower marketing automation and sales enablement tools like HubSpot, Outreach.io, SalesLoft, Salesforce, ...

CEO and Chief Product Officer

Sniper Intel

The idea behind Sniper Intel was originally born during my ex-PhD study on Decision Support Systems (switched to MSc to focus on Robust Choice) and also a few startups I was part of. We gradually focused on B2B sales space. We built Sniper Intel (working MVP) as a platform that aggregates and then distributes all information sales/marketing teams need, on the go.

Product Design and Lead

Softmax Data Inc.

Softmax Data is an advisory that focuses on providing services related to machine learning, scraping bots and data analytics. We work with internal teams to blueprint, build and deploy machine learning and data solutions to generate rapid results and long-term impact. Our client portfolio includes organizations across multiple industries such as government, agriculture, consumer electronics, software, and finance. Our team combines more than 60 years of industry experience serving Fortune 100 companies in building AI and data solutions from concept to production.

Partner

Steer Software

An online Priority Roadmap Tool that focuses teams on driving the business forward through priority-driven execution.

visit: www.steer.software

Product Manager

Pomegranate Housing Consultancy

Data Science consultant

TAGDit

TAGDit is a private database for any kind of group, where members can easily store and share information. The design is aimed to be as intuitive as an online search experience. This memory bank continuously grows, and as content is added, updated and rated, the data is refined.

CityPlus.io

City+ is a two-way mass communication platform for cities via messaging apps.  It started with the idea of how to help thousands of people in need after an earthquake. It requires a huge call center that is not feasible. We aimed to use Artificial Intelligence and automation technologies to enable city organizations to start and maintain an ongoing personalized communication with thousands to millions of people when it matters the most.

Product Manager & Innovator

HCTISS

Designed the product SEDA, a online sustainability decision support tool for construction industry.

Developed MVP and conducted user experiments and surveys

Hired and led the development team.

T.T. ITOK

At ITOK, we developed and implemented Geospatial Information System (GIS) in multiple cities. I managed several teams to conduct extensive data collection.

I worked closely with GIS experts to build and implement spatial databases and analytics services into multiple organizations.

I was responsible for communications with clients and trained local staff to utilize the solution in their processes and daily activities.

Project Manager

With Personize, HubSpot customers can create personalized sales sequences, marketing emails, and web pages for each lead—not just basic tokens or segmentation.

CRM systems don't provide this level of personalization.

Current CRMs mainly offer {​{First Name}} tokens, broad segmentation based on structured data, while 90% of customer data is unstructured, and A/B tests.

Unique content generation for each contact at scale remains a challenge for standard CRM tools.

AI SDR tools send “personalized” emails for common scenarios but lack the deep contextual understanding necessary for relevant and accurate personalization at scale. Unlike Personize, they don't have the qualifiers and reasoners required.

Users see us as a native alternative in HubSpot to tools like Clay or AI SDRs like Artisn, but we are 10x cheaper, more customizable, and more accurate.

✔ Customer stories:
◾ Outbound emails after deep account qualification and ICP match-making
◾ Personalized campaigns based on purchase history (e-commerce)
◾ Proposals and follow-ups dynamically generated from web form data and public search
◾ Localizing 60,000+ pages with region-specific info (B2B e-commerce)
◾ Cutting report-writing time from 20 hours to 4 (Commercial Real Estate)
◾ Account-level landing pages (B2B SaaS)
◾ AI research agents assisting technical sales (manufacturing, IoT, Oil & Gas)
◾ Re-engaging past customers and trial users with tailored campaigns

About me:
I bring 20+ years in data science, AI, and ML—leading go-to-market strategies and hands-on delivery for B2B solutions. I’ve worked directly with CMOs, VPs of Sales, and Founders to reimagine what’s possible with tech in sales and marketing.

◾ Deep domain expertise: Implemented 60+ RevOps and sales infra solutions since 2018, covering sales, marketing, and AI operations.

◾ Academic foundation: Dual MScs in Machine Learning and Visual Analytics (UBC & IUT), focusing on modeling complex consumer behavior and developing interactive decision support tools for billion-dollar projects.

◾ Product leadership: Led product development for a knowledge management platform (3 years)—built search, tagging, and knowledge graph features, central to our Unified Proactive Memory.

◾ Innovation track record: Designed and delivered city-wide analytics platforms (pre-2011), gaining experience in scaling data systems and supporting complex rollouts.

◾ LLM + AI: Expert in prompt engineering and deploying LLMs in real products.

◾ Team builder: Built and led AI and engineering teams, turning concepts into scalable products.

Hamed Taheri

Something shifted in the last six months.

Amazon, LinkedIn, Google, Microsoft, Salesforce, they're all building the same thing from different directions. And the thesis is identical:

AI agents that don't know how your organization works will eventually do something your organization wouldn't approve.

The race isn't to build smarter models anymore. It's to build the knowledge infrastructure that keeps agents aligned with the organizations they serve.

In Part 1, I break down how LinkedIn and Amazon are approaching this.

Two companies hit the same wall from opposite directions. LinkedIn's agents lacked organizational context. Amazon's agents had context but produced wildly inconsistent behavior. Both arrived at the same conclusion: structured, governed knowledge is the real bottleneck.

I share Part 2 tomorrow: Google, Microsoft, Salesforce, and what the pattern across all five reveals.

CEO at Personize.ai | technical product lead, agentic AI | memory, governance, enterprise adoption

View newsletter: Agentic Systems in Production

The biggest companies in tech are converging on the same conclusion: AI agents without organizational knowledge are a liability. Here's what they're e

Open article: The Race to Build Agentic Knowledge Bases - Part 1: LinkedIn, Amazon by Hamed Taheri • 6 min read

The Race to Build Agentic Knowledge Bases - Part 1: LinkedIn, Amazon

taplio

The Battle

Joshua Behar

Hamed Taheri

Followers

Posts

Engagements

Likes

Performance ScoreJoshua Behar's LinkedIn Performance

Performance ScoreHamed Taheri's LinkedIn Performance

Ready to grow your LinkedIn presence?

taplio

The Battle

Joshua Behar

Hamed Taheri

Followers

Posts

Engagements

Likes

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);}Joshua Behar's LinkedIn Performance

Performance ScoreHamed Taheri's LinkedIn Performance

Ready to grow your LinkedIn presence?

Performance ScoreJoshua Behar's LinkedIn Performance