E-Commerce Technology Platforms

Army launches G-TEAD Marketplace. Brigade commanders can buy counter-drone systems in 90 days. No RFPs. No committees. November 24, 2025. Global Tactical Edge Acquisition Directorate Activates Digital Procurement Platform. First vendor: Fortem Technologies with DroneHunter interceptors validated in Ukraine. Translation: Amazon for Army tech just went live. The disruption metrics. • Traditional acquisition: 2-3 years • G-TEAD timeline: 90 days max • Decision authority: Tactical commanders • Budget ceiling: None (draws from existing O&M) This isn't procurement reform. It's a procurement revolution. xTechCounter Strike wrapped in Germany. 15 competed. 5 won. Fortem's DroneHunter killed Russian Orlan-10s in Ukraine last month. Now any brigade commander orders them direct. No middlemen. No waiting. The eligibility gates. • CAGE/NCAGE code required • Master Business Agreement signature • Combat validation preferred • NATO partners included • Delivery in 90 days or contract void Three shifts crystallize. Theater beats Pentagon. Brigade commanders bypass acquisition bureaucracy entirely. Commercial beats custom. Ukraine-proven COTS dominates traditional programs. For contractors: Your competition isn't Raytheon anymore. It's whoever validated their tech in Bakhmut last week. Is your solution marketplace-ready or still chasing 5-year PORs? ---------- Like this content? Join our newsletter. Link located below my name 👆

In the emerging business of ‘social shopping,’ TikTok dominates the "social" and is spending big to improve the "shopping." Amazon dominates #ecommerce "shopping," and is working on the "social." TikTok made a name for itself in the U.S. as a viral video-sharing sensation. Now it’s trying to get its 150 million U.S. users to think of it as a shopping destination. Amazon, meanwhile, is trying new tactics to maintain its dominance in e-commerce. It has added social elements to its app to entice younger shoppers, and it is building up a network of influencers who hawk items on and off its website. As a result, the two companies are on a collision course as they vie for position in a huge market. Researchers at Insider Intelligence estimate social e-commerce will grow into a $100 billion market by 2025, from $67 billion this year. To succeed, each company will need to copy elements of the other’s success. TikTok, owned by Beijing-based ByteDance, wants customers to trust it as a safe and reliable place to buy products, the way many already trust Amazon. And Amazon is trying to persuade users to hang out on its app like they do on #socialmedia services. TikTok launched its #shopping feature, called TikTok Shop, last month and is currently selling about $7 million worth of products like hairbrushes, teeth-whitening tools and fall-themed sweatshirts with leaves and pumpkins every day in the U.S., with a goal of reaching $10 million a day by the end of the year. Amazon’s global online store sales—a measure of the products Amazon sells directly—was roughly $603 million a day last year. TikTok is spending heavily to build a logistics operation, poaching Amazon employees and trying to lure third-party sellers by offering them a bigger cut of sales than Amazon. More than 60% of Amazon’s retail sales come from third-party sellers. Amazon has spent years building up trust among consumers, thanks to a relentless focus on customer experience, including increasingly speedy deliveries and a lenient return policy. TikTok still has to earn that kind of trust. TikTok’s biggest advantage is its ubiquity in the U.S. The roughly two hours a day on average that U.S. users spend on TikTok leaves other companies jealous. On Amazon, U.S. customers spend an average of about 9.7 minutes a day. In a bid to keep users engaged for longer and to create shopping experiences that would appeal to younger users, Amazon has been integrating more social features into its app. One, called Inspire, shows users a TikTok-style feed of photos and videos featuring products customers can purchase. Amazon recently began to allow shoppers to upload content themselves on the Inspire tab and also added a “share” button that allows users to share the content and products they see. Customers have viewed over one billion Inspire posts since the feature launched last year. https://lnkd.in/eDc-6Jtm

I'm excited to share this impressive AWS case study featuring our customer Lion, a leading beverage company that successfully modernized its infrastructure running SAP applications through cloud adoption. Key Highlights: - 25% reduction in TCO - 70% faster data transfer speed - 100% preservation of SAP System Integrations - Enabled innovation with easier access to AWS’s AI services We in AWS Support are impressed by how organizations like Lion leverage cloud technology to drive business value. Their success story showcases the power of well-executed cloud migration strategies and the strength of the AWS-SAP partnership! Read the full case study : https://lnkd.in/gq7dZ3sm Planning your own large-scale migration or high-stakes launch? Learn how expert support can maximize success and minimize risk with AWS Countdown Premium.

When you query AI, it gathers relevant information to answer you. But, how much information does the model need? Conversations with practitioners revealed the their intuition : the input was ~20x larger than the output. But my experiments with Gemini tool command line interface, which outputs detailed token statistics, revealed its much higher. 300x on average & up to 4000x. Here’s why this high input-to-output ratio matters for anyone building with AI: Cost Management is All About the Input. With API calls priced per token, a 300:1 ratio means costs are dictated by the context, not the answer. This pricing dynamic holds true across all major models. On OpenAI’s pricing page, output tokens for GPT-4.1 are 4x as expensive as input tokens. But when the input is 300x more voluminous, the input costs are still 98% of the total bill. Latency is a Function of Context Size. An important factor determining how long a user waits for an answer is the time it takes the model to process the input. It Redefines the Engineering Challenge. This observation proves that the core challenge of building with LLMs isn’t just prompting. It’s context engineering. The critical task is building efficient data retrieval & context - crafting pipelines that can find the best information and distilling it into the smallest possible token footprint. Caching Becomes Mission-Critical. If 99% of tokens are in the input, building a robust caching layer for frequently retrieved documents or common query contexts moves from a “nice-to-have” to a core architectural requirement for building a cost-effective & scalable product. For developers, this means focusing on input optimization is a critical lever for controlling costs, reducing latency, and ultimately, building a successful AI-powered product.

In our journey of building personalized recommendations, we often debate when models should run in real-time vs. batch processing. It completely depends on use case, scalability, and latency that is acceptable. Let me try to simplify it so that you can explain it better to your management - 1) Real-Time Models – When Instant Personalization is Key. This flow is used when recommendations must be generated instantly based on a user’s current actions. Example Use Cases: "You May Also Like" – A user clicks on a product, and recommendations are generated dynamically. Personalized Home Page – When a user logs in, their recommendations are fetched in real time. Dynamic Offers – Based on recent user behavior, a discount or coupon is displayed immediately. This is how it can be implemented if using Amazon Web Services (AWS): 🔹 User Action → A user visits a webpage or clicks on a product. 🔹 API Gateway + Lambda → Triggers an API call to fetch recommendations. 🔹 Model Prediction (SageMaker Endpoint) → If no cached results exist, the model generates new recommendations. 🔹 DynamoDB / Redis Cache → First checks for recent recommendations to reduce latency. 🔹 Response to Frontend → Results are returned and displayed instantly. 2) Batch Processing – Precomputed Recommendations This approach is used when personalization can be precomputed, reducing the need for real-time execution. Example Use Cases: "Your Favorites" (Rule-Based Personalization) – If a user buys from X retailers frequently, precompute recommendations daily. Periodic Email / Push Notifications – Personalized product suggestions for email marketing campaigns. Homepage Personalization (Static User Preferences) – Daily updates to improve page load speed. This is how it can be implemented: 🔹 Daily / Weekly Training Jobs (Glue, SageMaker, EMR) → or you can use dedicated EC2 & Jenkins to process large amounts of data and update recommendations. 🔹 Updated Recommendations Stored (DynamoDB, Redis) 🔹 Precomputed Recommendations Served via API / CloudFront So, if recommendation changes dynamically basis user session, use real time. For predictable updates use batch. Infact, one can use hybrid approach also - Cache precomputed results and fall back on real-time inference when needed. #recommendation #n=1personalisation #datascience #data

AI Agents Don’t Buy Seats—Why Your Pricing Should Follow Suit In the past 12 months, a clear pattern has emerged: as AI systems replace manual effort with automated intelligence, pricing structures tied to “seats” no longer reflect the value customers receive. Pricing models have surfaced as a hot topic with every portfolio company at Mosaic Ventures and is top-of-mind for nearly every founder building applied-AI products. When one person and an AI agent can outperform an entire legacy team, charging per user starts to feel arbitrary; what matters is how much business impact the product delivers. Founders are experimenting with three broad approaches: 1. Usage-metered plans that bill against tokens, API calls, or minutes of inference time. These create a direct bridge between consumption and margin and nudge teams to track cost from day one. 2. Outcome-based pricing that charges per lead booked, ticket resolved, or document drafted—tying revenue to measurable results. It’s the software analogue of value-based care. 3. Hybrid “starter bundle plus runway” tiers: a predictable monthly fee with a healthy allowance of AI credits, then pay-as-you-go beyond that. This balances budget certainty for customers with upside capture for the vendor. Across our portfolio, a few design principles keep showing up: 1. Anchor on a metric the customer already tracks. If your product shortens sales cycles, price per opportunity accelerated—not per login. 2. Bundle enough volume to eliminate credit anxiety. No one wants to ration prompts. 3. Expose real-time usage. Transparent dashboards prevent bill shock and build trust. 4. Instrument cost early. Metering and billing belong in the product backlog, not the finance queue. 5. Plan for non-linear jumps. When a model upgrade multiplies compute, re-grade tiers before your gross margin does it for you. AI’s promise is to shift human effort from repetitive execution to higher-order creativity. If our pricing still counts bodies instead of business results, we undermine that promise. The companies that map price to outcomes—while keeping the buying experience refreshingly simple—will capture the most upside. I’d love to hear how others are managing the move from seats to usage and outcomes. What’s working, what still feels messy, and where do you see the biggest opportunities to innovate on pricing? #appliedAI #pricing #startups

Pricing model is moving from subscription to usage/volume-based pricing for AI products. With traditional SaaS, the cost of operating and maintaining the software is relatively fixed given a certain scale of users. No matter if the user is doing 1 action a week or 100 actions a day, your server and infra cost remains relatively unchanged. However, with AI and LLM models, the cost of serving a power user who sends 200 messages a day, vs the cost of serving a casual user who sends 1 message a day is vastly different. This is because the infra cost of running inference is directly correlated with the number of user messages and tokens. For companies building on top of LLMs, the bulk of the operating cost is API cost thst is also directly tied to usage. You see ChatGPT trying to address this disparity by introducing the 2 tier plan ($20 vs $200). And Claude is facing constant complains on the usage limit. Newer companies like Devin and Cursor has taken the right step to introduce flat volume based pricing. You just pay $500 dollars for X amount of compute time, or $20 dollars for Y amount of messages. I believe volume based pricing is the future.

Today, the demo version of #DOT-Chain #Defence — a digital system for supplying drones to the frontline — was officially presented. In essence, it’s a marketplace for drones and robotic systems, where: • #Military units can see what’s actually available — directly, without layers of approvals or delays. • #Manufacturers can showcase their developments on the platform and supply equipment transparently, responding to real and current demands. • And logistics are reduced from months to weeks — and that’s the key breakthrough. It may sound simple. But in practice — it’s a major step forward. Until now, the process looked something like this: a unit submitted a request, it moved up the chain of command, and somewhere along the way, things would disappear or suddenly appear out of nowhere. Complete lack of transparency. Now, military personnel see their options clearly. No more Excel spreadsheets. No back-channel calls. No guessing what might be needed next month. Just accurate information and direct choices — think of it like Rozetka.ua (Ukraine’s Amazon but for defence. Of course, there are still questions: Who exactly will have access? How will selection and vetting be handled? How will adjustments be made when equipment needs customization for specific tasks? But the key point remains: for the first time, we’re seeing a living, working system that directly connects supply and demand. Clearly. Quickly. Humanely. At the Ukrainian Council of Defence Industry - Рада Зброярів we fully support this initiative. Because when manufacturers stop working blindly, and the military stops waiting for months — efficiency wins. Not bureaucracy. And in this case, time is not just about cost. It’s about tactical advantage. And opportunities we cannot afford to lose.

Units will be able to exchange e-Points for drones, EW systems, and UGVs on the Brave1 Market, and receive UAVs faster Now fully digitized, the bonus-based drone distribution system will operate more efficiently. Service members can redeem confirmed combat kills for a broader range of high-impact technologies. The entire process — from order placement to delivery — will take no more than a few weeks. What’s changing: 🔸Units can now use points earned for confirmed enemy kills to request technologies directly via the Brave1 Market https://lnkd.in/gCbieCHg, using their credentials through the DELTA system. 🔸The e-Points can be exchanged for codified UAVs, electronic warfare (EW) systems, UGVs, and non-codified FPV drones. This will allow units to choose precisely what they need from various technologies to carry out their missions effectively. 🔸The MoD’s Defence Procurement Agency will process orders through DOT-Chain Defence. 🔸All documentation, from contracts to delivery receipts, will be generated online. There will be no paperwork, delays, or manual form-filling. In addition, BRAVE1 will now handle both the verification of enemy kills and the awarding of points. This will ensure accurate tracking and statistical reporting and enable real-time analysis of which technologies are proving most effective on the battlefield. The Ministry of Digital Transformation will continue shaping the program's policy, while the Defence Procurement Agency will be responsible for timely and efficient procurement and delivery. Units can place orders under the new model starting in August 2025, using combat points earned in June or any unused points from earlier periods. The Army of Drones.Bonus system has become a game changer at the front. Troops gained an extra source of motivation, and the state gained valuable battlefield data for informed decision-making. In just over 10 months, more than 400 units have joined the program, roughly 95% of all drone-operating units across the front line. In that time, the number of enemy systems destroyed has doubled. Thousands of enemy assets, including tanks, UAVs, EW systems, depots, air defense systems, artillery, and howitzers, are destroyed every month. We constantly monitor how the system works and make real-time updates, so you don’t have to deal with red tape and can stay focused on defense and destroying the enemy. Thank you to the service members doing extraordinary work every single day.

Feals built a $15M+ business with 15K customers in 8 months because they cracked the code on turning anxious browsers into lifetime subscribers Let’s break down their subscription playbook THE SAMPLING PSYCHOLOGY BREAKTHROUGH Most CBD brands force customers to commit to full bottles without knowing what dosage works. Feals created “The Flight”, a $20 sampler with 3 strengths of: 👉 40mg 👉 80mg 👉160mg Customers discover their perfect dose over three days, then get locked into a membership saving 30% for life. The genius move was eliminating CBD’s biggest friction point, dosage uncertainty. DATA DRIVE FOUNDATION Before Feals, the founders ran a proof of concept, “Vitalife”, where they: 👉 Tested every CBD format 👉 Sold 6K+ customers 👉 Generated $1M revenue They built data first, then created Feals around proven winners. CREATIVE DISTRIBUTION WHEN PLATFORMS FAIL Facebook and Google banned cannabis ads, so Feals went where conversations happen, like:  👉 Podcast mid-rolls on “Pod Save America” and “Girls Gotta Eat” 👉 First CBD hotline for dosage consultations 👉 Vanity links for attribution tracking CALM BRANDING IN A CHAOTIC MARKET While competitors screamed about benefits, Feals whispered confidence with: 👉 Soft beige tones and serif fonts 👉 "Clear mind, calm body" messaging focused on feelings 👉 Educational content over product-pushing 👉 Looked like wellness therapy, not supplements THE RESULTS Results that validated the strategy included:  1️⃣ 600% revenue growth in 8 months 2️⃣ 70%+ subscription retention 3️⃣ Backed by Glossier investors Feals proved that in wellness DTC, sampling beats advertising and subscription beats one-time sales. The calm brand won by making CBD feel safe enough to try and convenient enough to continue.