Sam Shawki

Forward deployed engineers are consultants in hoodies — but with tech multiples 🥷💸 FDEs (forward deployed engineers) are a great GTM hack. And right now, the revenues they bring in are often counted as ARR — and get the corresponding multiple. 😅 You embed engineers with customers, do a bunch of custom work, make sure they leverage your full suite and get tightly integrated into their pipelines. The result? Consulting-style delivery — but recurring revenue optics. OpenAI’s FDE engagements start at $10M. Rumors say they’ve landed a $200M FDE deal with the Pentagon. ...I’ve also heard of several London PE firms bringing OpenAI in to deploy FDEs across their portfolios to automate anything that moves. But let’s be honest — this is services revenue in SaaS clothing. High-touch, low-margin, labor-intensive work, bundled with software or models. It’s a brilliant model though! You get sticky integrations, higher retention, and short-term revenue that (for now) still gets valued like software. Wonder for how long it will go on? The lines between “product” and “professional services” are already blurry. We might see a reckoning in how hybrid models are valued. But for now, the best players are still pulling it off. 🔥 #FDE #AI #OpenAI #VC #Valutations

26

7 Comments

Looking forward to NANOG 96 in San Francisco & Damien Garros and Wim Henderickx's talk on the Network Automation Forum automation framework. A lot of organizations still struggle to move beyond ad-hoc scripts, brittle workflows, and unclear architectural models. The NAF Automation Framework is a practical, engineer-friendly approach to building automation on a modular reference architecture that was designed by practitioners for practitioners. The building blocks of the framework that the talk will cover include: 💠 Intent 💠 Observability 💠 Collector 💠 Executor 💠 Orchestrator 💠 Presentation Wim and Damien will also cover automation best practices: 💠 Declarative design 💠 Idempotent execution 💠 Diff-based workflows 💠 Version-controlled changes 💠 Safe failure handling 💠 Human-friendly interfaces. Worth checking out. #networkautomation

28

4 Comments

DDN is entering its next phase — the industrialization of AI at global scale. I’m excited to announce Mohsen Moazami as Vice Chair. Mohsen helped scale Groq into a category-defining AI platform and played a key role in its landmark $20B strategic engagement with NVIDIA. The AI industry has reached a clear inflection point. In Jensen’s words, compute is no longer the constraint, DATA is. DDN has been built for this moment. We operate the data foundation behind many of the largest AI factories, sovereign AI platforms, and mission-critical AI deployments in the world. Mohsen’s global operating and capital markets experience and reach strengthens our ability to scale DDN as the foundational platform of the World's AI economy. To the Stars! #IPO #PublicMarkets #CapitalMarkets #SovereignWealthFunds #SovereignAI #AIInfrastructure #AIEconomy #PlatformCompany #LongTermValue #NVIDIA #DDN

143

9 Comments

In edge device development, five decisions made in the first few weeks determine whether a product ships on time — or gets stuck in redesign loops that burn months and budget. After two decades of helping companies move from concept to production, we've seen these same five inflection points come up every time: compute core, power architecture, firmware stack, connectivity, and manufacturability. The teams that get to market fastest aren't necessarily the ones with the biggest budgets. They're the ones who make these calls early, decisively, and with the right experience behind them. Our team wrote down everything we've learned. Worth a read if you're building for the edge. https://ow.ly/sUPw50YcYUj #EdgeAI #EmbeddedSystems #ProductDevelopment #EdgeComputing

34

1 Comment

MCP servers are integral to agentic architectures, enabling AI agents to integrate seamlessly with external tools and data sources. However, this integration requires special attention to key security challenges:​ 🔒Authentication and Authorization: Verify identities and control access to ensure only authorized entities interact with MCP servers.​ 💡Real-time Context Validation: Ensure interactions are appropriate and authorized in real-time.​ 🛡️API-Level Introspection: Monitor and analyze API calls for anomalies.​ 🚨Threat Detection for Autonomous Behaviors: Identify and mitigate risks associated with AI-driven operations.​ Long live secure MCP servers. #AgenticArchitecture #AISecurity #MCPServers #Centerofexcellence #evolutionportfolio

29

4 Comments

Despite the many benefits of Data #CleanRooms they may turn participating parties into Data Brokers as per California’s DELETE Act if they happen to bring 3rd party data (no direct relationship) into the platform, triggering a registration and tech alignment requirement (and hefty fines if dates are missed). (Presentation by California Privacy Protection Agency and Ben Isaacson on the DELETE ACT and the new #DROP platform that integrates all data brokers for consumers to place a single request.) Privacy Law Section of the California Lawyers Association. #AdTech

57

5 Comments

Cellular cores have fallen in price from starting at 6 or 7 figures, to just a few thousand or even a few hundred dollars in a matter of a few years, thanks to the work of Open5gs, Althea team, Hawk Networks, Inc. with KeyLTE/5G, University of Washington, and many others. Came across the post in an ISP industry group and it showcases well the change in pricing and the unlock to building in cellular that is. This was also fueled by accessible spectrum in CBRS, with support from groups like OnGo Alliance. While spectrum remains a challenge, innovative partners like Globalstar chart a path forward. With sim orchestration aligned thanks to companies like Conekt.ai, radio hardware remains one of the last challenges to complete an affordable cellular stack. Behind the scenes, there was a quiet group of builders, researchers, universities and advocates that recognized the direct to device revolution before it was a household reference. This article on Ulukhaktok illustrates this history of empowering operators and communities of all sizes to leverage and build sovereign and resilient comms and gives an exciting glimpse of what's to come in cellular connectivity. https://lnkd.in/gsiXKjTX

24

3 Comments

Last week we shared that ZaiNar had emerged from nine years of stealth. Today we’re sharing what we built. We’re announcing commercial availability of the first 5G location system that delivers sub-10cm accuracy using only the connectivity signals devices already transmit. Ranges up to 1.5 kilometers. Zero additional battery drain. No device software required. Here’s why this is a big deal for the telecom industry: carriers have invested billions in 5G infrastructure globally. With ZaiNar, those networks can now deliver highly accurate positioning services across every connected device - dramatically increasing the utility and ROI of those investments. Location services can be worth more than the connectivity itself. Our technology uses SRS - the connectivity signal every device transmits to maintain its network connection - and processes it at the network level. The result is sub-10cm positioning that works indoors, outdoors, through walls, on as little as 10 MHz of spectrum. This turns 5G from a faster pipe into genuine infrastructure for Physical AI - and it’s deployed and operating today across healthcare, construction, logistics, and smart cities. With MWC next week, we have much more to share. https://lnkd.in/ghVhYXvt #ZaiNar #5G #PhysicalAI #noGPS #MWC2026

137

29 Comments

As AI workloads eat up global computing supply, DRAM prices are surging like never before—creating major challenges for teams trying to plan out their data center spend. One of my portfolio companies, Mext, is addressing the largest cost component in the datacenter: server memory (DRAM). From being on the board of Redis for the last twelve years, I’ve learned a lot about the various aspects of this problem and approaches to solving them. Mext found a breakthrough, using new AI techniques, for dramatically reducing the amount of server-DRAM required to run applications, all while maintaining performance. It intelligently manages the server’s memory, keeping hot pages (i.e., those in use by applications) in fast DRAM and offloading cold pages (i.e., those less used) to a much less expensive memory tier (e.g., NVMe Flash). Key to the approach is ensuring cold memory pages that are about to be accessed by an application are back in DRAM before the application needs them or notices that they were gone. This is done without modifications to the application or the OS – so it can run in the cloud or on-premise. Mext thus allows applications to either run using less DRAM or keep their DRAM footprint but do more with it. They're hosting a webinar along with Fred Weber (former CTO of AMD) on Jan 22nd at 10am pacific time. You can register here: https://lnkd.in/gXe9GCeE

153

3 Comments