Kobi Levi

AI unicorn Gong to add 100 jobs following revenue surge. The Israeli company has hit a $300M run rate and is growing its global presence with new offices, in addition to significantly increasing its local workforce. https://lnkd.in/eJ6GM93M

114

1 Comment

MetalBear, an Israeli startup that created the open-source Kubernetes tool mirrord, has raised $12.5 million in Seed funding. The company aims to solve a major challenge in software development: testing code in realistic cloud environments. The funding round was led by TLV Partners and included investments from TQ Ventures, MTF, Netz Capital, and angel investors such as Sentry co-founder David Cramer and OpenTelemetry co-creator Ben Sigelman. Eyal Bukchin | Aviram Hassan 🔗 Read full article link in comment #metalbear #tlvpartners #startupnews #funding #israel #businessupdate

107

2 Comments

Another milestone for Israeli cybersecurity and YL Ventures: Aim Security, led by founders Matan G. and Adir Gruss, has been acquired by Cato Networks! From their $10M seed round in 2023 through their $18M Series A, we had the privilege of working closely with the team as they built one of the fastest growing cybersecurity companies in Israel, and the category leader of AI Security. From day zero, Aim set out with a bold vision: to give enterprises a way to adopt AI securely and responsibly. In two years, Matan, Adir, and their exceptional team turned that vision into a reality, building a company trusted by some of the world’s largest organizations. This achievement demonstrates the power of brilliant founders, strong teams, and conviction from inception. For YL Ventures’ $400M Fund V, this is a significant outcome and another marker of Israel’s unique ability to produce companies that matter on a global stage. https://lnkd.in/erHnKmfN

133

1 Comment

Last week, I had the privilege of facilitating a panel in Midtown Tel Aviv with the Co-founders & CEOs of Silverfort (Hed Kovetz ), Oligo Security (Nadav Czerninski) and CTO of Torq (Leonid Belkind) hosted by friends at Greenfield Partners. This was easily one of the most intellectually stimulating conversations I’ve had this year, exploring the intersection of runtime security and AI agents. Key discussion highlights: The common thread we explored is that runtime security is becoming the battleground for securing AI agents. ▪️ Oligo secures what’s running in the application layer. ▪️ Silverfort secures the identity layer on who/what is accessing resources in real time. ▪️ Torq orchestrates the operational glue to act on these triage at scale. We see the biggest AI agent risk opportunities for practitioners around: ▪️ Shadow AI adoption: Teams deploying agents without security oversight. ▪️ Unintended actions: Agents causing data loss or disruption without malicious intent. ▪️ Exploitation: Attackers targeting agent vulnerabilities or tricking them into revealing data. ▪️ Autonomy vs. control: Challenge of applying least privilege and dynamic access controls. Key themes: ▪️ Silverfort (Identity Security) emphasized that identity is the most crucial layer, with AI agents. The opportunity is to enforce identity security in real time, dynamically deciding what agents can and cannot access. Yet the risk is that, unlike humans, agents cannot be MFA-challenged, meaning that without the right guardrails they may unintentionally expose sensitive data or perform actions at machine speed. We need runtime guardrails. ▪️ Oligo (Runtime App Sec) showed how runtime visibility has become indispensable in an age where AI agents drive dynamic workloads. The opportunity lies in helping organizations truly see what’s happening inside their applications, but the risk is clear: rushed adoption of AI often introduces vulnerabilities that static, shift-left approaches will never catch. ▪️ Torq (Hyperautomation & AI for Security Ops): We see Torq as the connecting glue for all companies and units in an agentic ecosystem. We highlighted how AI-driven automation will drastically reduce response times, orchestrating investigations and containment in minutes rather than days. We're all excited about the future opportunities that AI agents bring within Identity, SecOps, and Cloud App runtime. Together, these perspectives form a runtime security fabric, one that will define how enterprises defend themselves in an era where AI agents are both the greatest opportunity and the fastest-moving risk. *** Massive thank YOU to Greenfield Partners, Josh Trup, Nir Goldstein, Meir Cohen, Ortal Sasson Hendin, Raz Mangel and everyone for bringing this together and making it work! (The best spot in Tel Aviv!)

215

9 Comments

What do soccer, chess, and real time AI analysis have in common? 🤔 At Google Cloud AI Tel Aviv, Google Cloud Israel’s Managing Director Boaz Maoz and Head of Cloud Marketing Elit Ben Basat Nuriel gave us a taste of the AI Playground where the latest tech met serious fun. 𝑺𝒑𝒐𝒊𝒍𝒆𝒓: Gemini Flash models running live game analysis at the AI Playground, providing instant advice and winning moves that were even more exciting than the games themselves. Watch how the AI Playground brought AI to life in ways you don’t see every day ⤵️

65

4 Comments

Crusoe is acquiring Israeli startup Atero AI in a deal valued at approximately $150 million. The acquisition involves a cash-and-stock transaction, and Atero's 25 employees will become Crusoe's development center in Israel. Atero, founded in July 2024, specializes in GPU management technology for AI workloads, and has raised $12 million in a seed round. Alon Yariv, CEO of Atero, said joining Crusoe is the fastest path to achieving their mission, while Chase Lochmiller, CEO of Crusoe, highlighted the benefits of Atero's expertise in high-performance computing infrastructure. Read more: https://lnkd.in/eKEgVEPe 📰 Subscribe to the weekly Silicon Brief Newsletter: https://lnkd.in/ejfzg92J #ai #artificialintelligence #ainews #aifunding #aichips #gpu

100

2 Comments

How hard can it be to build an AI Coding Assistant???? That was the question that motivated me to create the workshop I delivered today as part of the Microsoft Israel internal AI Days. The session, titled "Creating AI Applications and Agents with C# and Semantic Kernel", was a hands-on deep dive into building intelligent agents using modern AI tooling and .NET. For those curious or eager to experiment, I’ve uploaded all the materials to my GitHub: 🔗https://lnkd.in/d9WhWivk Inside you'll find: - Jupyter notebooks with the full workshop content - Step-by-step exercises that guide you in building your own AI Coding Assistant (think Cline / Roo / GitHub Copilot) just like you see in the attached video - A VS Code extension that activates the assistant directly in your editor Hope you find it useful—and if you try it out, I’d love to hear your thoughts!

47

2 Comments

“𝗘𝘃𝗲𝗿𝘆 𝗱𝗲𝗹𝗮𝘆 𝗶𝘀 𝗮 𝗱𝗮𝘁𝗮 𝗽𝗼𝗶𝗻𝘁 𝘄𝗮𝗶𝘁𝗶𝗻𝗴 𝘁𝗼 𝗯𝗲 𝗻𝗼𝘁𝗶𝗰𝗲𝗱. 𝗕𝘂𝗶𝗹𝗱𝗼𝘁𝘀 𝗺𝗮𝗸𝗲𝘀 𝘀𝘂𝗿𝗲 𝘆𝗼𝘂 𝘀𝗲𝗲 𝗶𝘁 𝗯𝗲𝗳𝗼𝗿𝗲 𝗶𝘁’𝘀 𝘁𝗼𝗼 𝗹𝗮𝘁𝗲.” - Ori Silberberg, VP of Engineering, Buildots Too often, project setbacks aren’t caused by big surprises - they stem from invisible inefficiencies that stay hidden until it’s too late. On the Machine Learning Podcast, hosted by Tobias Macey, Buildots’ VP of Product & Engineering, Ori, joined the conversation to explore how AI and computer vision are helping construction teams track progress, identify root causes of delay, and forecast risks across projects without adding tasks for site crews. Already used across 1.4 billion sq ft of projects, from airports to data centers, Ori explains how Buildots AI helps teams reduce waste and take back control. 🎧 Catch the full conversation here: https://lnkd.in/eAmQpvvN #Construction #ConstructionAI #Constructiontech

45

🔷  4 𝐏𝐢𝐥𝐥𝐚𝐫𝐬 𝐨𝐟 𝐃𝐚𝐭𝐚 𝐒𝐜𝐢𝐞𝐧𝐜𝐞 𝐌𝐚𝐬𝐭𝐞𝐫𝐲  𝐓𝐨 𝐭𝐡𝐫𝐢𝐯𝐞 𝐢𝐧 𝐃𝐚𝐭𝐚 𝐒𝐜𝐢𝐞𝐧𝐜𝐞, 𝐛𝐚𝐥𝐚𝐧𝐜𝐞 𝐭𝐡𝐞𝐬𝐞 𝐜𝐨𝐫𝐞 𝐩𝐢𝐥𝐥𝐚𝐫𝐬 - 𝐞𝐚𝐜𝐡 𝐞𝐬𝐬𝐞𝐧𝐭𝐢𝐚𝐥, 𝐧𝐨𝐧𝐞 𝐨𝐩𝐭𝐢𝐨𝐧𝐚𝐥: 1️⃣ Computer Science → Algorithms, ML models, scalable code → Tools: Python, SQL, Git 2️⃣ Math & Statistics → Probability, linear algebra, hypothesis testing → Tools: NumPy, SciPy, R 3️⃣ Communication & Visualization → Turn raw data into clear, actionable stories → Tools: Tableau, Power BI, Matplotlib 4️⃣ Domain Knowledge → Industry context fuels meaningful insights → Tools: Excel, Google Analytics, Salesforce 🌟 Follow the AIKaDoctor (Free AI & Data Science Resources)  channel on WhatsApp: https://lnkd.in/dCTCEKKc 📌Follow Habib Shaikh For more such content.

53

12 Comments

𝗘𝘅𝗼𝘀𝘁𝗲𝗹𝗹𝗮𝗿.𝗮𝗶 𝗵𝗮𝘀 𝗿𝗲𝗮𝗰𝗵𝗲𝗱 𝗧𝗶𝗺𝗲𝘀 𝗦𝗾𝘂𝗮𝗿𝗲—𝗮𝗻𝗱 𝘄𝗲’𝗿𝗲 𝗻𝗼𝘁 𝘀𝘁𝗼𝗽𝗽𝗶𝗻𝗴 𝗮𝗻𝘆𝘁𝗶𝗺𝗲 𝘀𝗼𝗼𝗻. I watched the Exostellar logo flash across one of the biggest stages in the world. Times Square. To be honest, it was a surreal moment—not because of the lights, but because of how far we’ve come. From early brainstorms and late-night debugging sessions, to now being on the path to reshaping how AI infrastructure runs. We’re not just another infra startup—we’re the future of AI Infrastructure Orchestration As AI workloads scale, the industry needs more than raw GPU power—it needs smarter orchestration. That’s what we have built:  • Centralized orchestration for GPUs across multi-cluster, hybrid environments  • Dynamic optimization to eliminate idle silicon and infrastructure waste  • Secure, chip-agnostic multi-tenancy at the container level  • Enterprise-grade segmentation and visibility controls for teams and tenants We’re making it seamless for AI platforms to run anywhere, with better performance, lower cost, and no vendor lock-in. Our mission? Become the default layer for AI infrastructure orchestration—just like Kubernetes became for containers. Huge thanks to the Exostellar team, Nayan Lad, Yuchen Fama, Zhiming Shen, Mahsa Almaeenejad for making this journey unforgettable. We even captured some bloopers during the shoot—drop a comment if you want to see them 😄 Check out our brand new website to learn more and transform your infrastructure ROI - www.exostellar.ai #Exostellar #TimesSquare #AIInfrastructure #GPUOrchestration #StartupMilestone #FutureOfAI #InfraThatWorks #AWSSummit Onur Aksoy Rohan Prakash Christopher Edgar Dan M.

216

39 Comments

🧬 𝙎𝙩𝙧𝙪𝙜𝙜𝙡𝙞𝙣𝙜 𝘼𝙡𝙤𝙣𝙚? 𝙃𝙤𝙬 𝙩𝙤 𝙋𝙧𝙤𝙫𝙚 𝙈𝙤𝙡𝙚𝙘𝙪𝙡𝙖𝙧 𝘿𝙮𝙣𝙖𝙢𝙞𝙘𝙨 𝙎𝙠𝙞𝙡𝙡𝙨 𝙒𝙞𝙩𝙝𝙤𝙪𝙩 𝙖 𝙋𝙄 𝙤𝙧 𝙇𝙖𝙗 𝘼𝙘𝙘𝙚𝙨𝙨 𝘔𝘢𝘯𝘺 𝘴𝘵𝘶𝘥𝘦𝘯𝘵𝘴 𝘧𝘦𝘦𝘭 𝘴𝘵𝘶𝘤𝘬 𝘣𝘦𝘤𝘢𝘶𝘴𝘦 𝘵𝘩𝘦𝘺 𝘥𝘰𝘯’𝘵 𝘩𝘢𝘷𝘦 𝘢 𝘴𝘶𝘱𝘦𝘳𝘷𝘪𝘴𝘰𝘳, 𝘗𝘐, 𝘰𝘳 𝘭𝘢𝘣 𝘢𝘤𝘤𝘦𝘴𝘴 𝘵𝘰 𝘨𝘶𝘪𝘥𝘦 𝘵𝘩𝘦𝘮. 𝘽𝙪𝙩 𝙝𝙚𝙧𝙚’𝙨 𝙩𝙝𝙚 𝙩𝙧𝙪𝙩𝙝: You don’t need a supervisor to validate your skills. Recruiters and PIs don’t hire based on who approved your project - they hire based on whether you can demonstrate real outputs. 𝙈𝙤𝙡𝙚𝙘𝙪𝙡𝙖𝙧 𝘿𝙮𝙣𝙖𝙢𝙞𝙘𝙨 (𝙈𝘿) simulations are a perfect way to prove this, even if you’re learning solo. Here’s how to build a portfolio-ready MD project step-by-step ⬇️ 📌 𝗦𝘁𝗲𝗽 𝟭: Choose Your Protein • Get a structure from the Protein Data Bank (PDB) → https://www.rcsb.org • Or predict one with AlphaFold if no structure exists. • Pick something connected to disease biology for higher impact. 📌 𝗦𝘁𝗲𝗽 𝟮: Prepare the System • Use GROMACS or AMBER (free/open-source MD tools). • Add hydrogens, solvate in a water box, and neutralize with ions. • This setup ensures biological realism. 📌 𝗦𝘁𝗲𝗽 𝟯: Define Parameters • Select a force field (e.g., CHARMM, AMBER99SB). • Set simulation time (start with 5–10 ns for practice). • Save configuration files for reproducibility. 📌 𝗦𝘁𝗲𝗽 𝟰: Run the Simulation • Perform energy minimization, equilibration, and production runs. • Even short runs generate enough data to analyze. 📌 𝗦𝘁𝗲𝗽 𝟱: Analyze Results • Track RMSD, RMSF, hydrogen bonds, radius of gyration. • Visualize trajectories using VMD or PyMOL. • Export graphs/plots to highlight stability and interactions. 📌 𝗦𝘁𝗲𝗽 𝟲: Package the Project • Create a clean workflow diagram. • Include graphs + snapshots of the simulation trajectory. • Write a short case study showing what you learned. 🎯 𝗪𝗵𝘆 𝗧𝗵𝗶𝘀 𝗠𝗮𝘁𝘁𝗲𝗿𝘀 ✅ Adds independent research experience without a lab ✅ Shows recruiters your ability to handle MD simulations end-to-end ✅ Creates tangible outputs (plots, graphs, visuals) for your portfolio 🔥 This is exactly what we teach inside 𝙑𝙞𝙨𝙞𝙗𝙞𝙡𝙞𝙩𝙮 𝘼𝙘𝙘𝙚𝙡𝙚𝙧𝙖𝙩𝙤𝙧 - how to turn solo projects into recruiter-ready case studies that boost your visibility. 💬 𝘞𝘢𝘯𝘵 𝘵𝘰 𝘱𝘳𝘰𝘷𝘦 𝘺𝘰𝘶𝘳 𝘴𝘬𝘪𝘭𝘭𝘴 𝘦𝘷𝘦𝘯 𝘪𝘧 𝘺𝘰𝘶’𝘳𝘦 𝘸𝘰𝘳𝘬𝘪𝘯𝘨 𝘢𝘭𝘰𝘯𝘦? 𝘿𝙈 𝙢𝙚. 🎁 𝗚𝗶𝘃𝗲𝗮𝘄𝗮𝘆 𝗔𝗹𝗲𝗿𝘁 Every week, I’ll pick one of the best comments and create a 𝙁𝙍𝙀𝙀 𝙥𝙤𝙧𝙩𝙛𝙤𝙡𝙞𝙤 𝙬𝙚𝙗𝙨𝙞𝙩𝙚 tailored to that person’s info - 𝘱𝘭𝘶𝘴 6 𝘮𝘰𝘯𝘵𝘩𝘴 𝘰𝘧 𝘧𝘳𝘦𝘦 𝘩𝘰𝘴𝘵𝘪𝘯𝘨 & 𝘥𝘰𝘮𝘢𝘪𝘯! ⏰ 𝙒𝙞𝙣𝙣𝙚𝙧’𝙨 𝙣𝙖𝙢𝙚 𝙬𝙞𝙡𝙡 𝙗𝙚 𝙖𝙣𝙣𝙤𝙪𝙣𝙘𝙚𝙙 𝙚𝙫𝙚𝙧𝙮 𝙎𝙪𝙣𝙙𝙖𝙮 𝙖𝙩 12 𝙋𝙈.

18

𝐈𝐬𝐫𝐚𝐞𝐥 𝐜𝐚𝐧 𝐨𝐰𝐧 𝐚𝐧𝐲𝐭𝐡𝐢𝐧𝐠 𝐢𝐭 𝐰𝐚𝐧𝐭𝐬 𝐢𝐧 𝐀𝐈.   That’s the bold perspective from our own Eyal Niv, Managing Partner at Pitango First, in CTech by Calcalist’s VC AI Survey.   Israel is already active across the entire AI stack, from infrastructure to applications, and has the talent, vision, and ambition to lead globally. As Eyal explains, the future will belong to 𝐀𝐈-𝐧𝐚𝐭𝐢𝐯𝐞 𝐚𝐧𝐝 𝐀𝐈-𝐫𝐞𝐬𝐢𝐥𝐢𝐞𝐧𝐭 𝐬𝐭𝐚𝐫𝐭𝐮𝐩𝐬 with defensible moats built on unique datasets, not just “AI wrappers.”   At Pitango, we’ve seen this firsthand across our portfolio: companies like AI21 Labs, DriveNets, AA-I, PsiQuantum, Quantum Source, Volumez and more are proving that Israel’s edge in deep tech and AI infrastructure is only getting stronger.   Bottom line: What makes a lasting company? Unique data, defensibility, and founders willing to build for the long term. Where does Israel fit? Anywhere it chooses.   Read the full interview with Eyal here: https://lnkd.in/esm_SVP5   #Pitango #PitangoFirst #Ctech

26

1 Comment