Cosimo Spera

New release from Packt Publishing… “Building Business-Ready Generative #AI Systems — Build Human-Centered Generative AI Systems with Context-Aware Agents, Memory, and LLMs for the Enterprise” See it at https://amzn.to/3Jdcio5 𝓚𝓮𝔂 𝓕𝓮𝓪𝓽𝓾𝓻𝓮𝓼: 🔵Build an adaptive, context-aware AI controller with advanced memory strategies 🟢Enhance GenAISys with multi-domain, multimodal reasoning capabilities and Chain of Thought (CoT) 🟠Seamlessly integrate cutting-edge OpenAI and DeepSeek models as you see fit

14

3 Comments

AI operating system startup ProductNow has raised $6 million in new funding to launch its software stack for product, program, and strategy teams. ProductNow is building an intelligent, centralized AI system to replace the fragmented mix of documents, decks, and siloed tools that modern product organizations use. The agentic platform is designed to turn high-level business intent directly into coordinated execution, extending AI-driven acceleration beyond engineering to the entire cross-functional workflow, including product management, UX/design, and marketing. The seed round was led by SIERRA Ventures LP, with participation from Parameter Ventures and senior operators from leading cloud, consumer, and AI companies. The funding will be used to launch its software stack and build out a system of leverage that augments teams across the entire organization. Congratulations to founder and Chief Executive Tript Singh Lamba and the ProductNow team. SiliconANGLE: https://lnkd.in/dK-yShvw #ProductManagement #AI #OperatingSystem #SaaS #FutureOfWork #SeedFunding

10

1 Comment

Great conversation today with Preethi CN, Director of Agentic AI at Amazon Web Services (AWS). Her message was blunt: most AI agents fail in production not because of the models, but because they lack real infrastructure — feedback loops, observability, controls, and the rigor needed for production-grade autonomy. Agentic AI only works when the infrastructure is built for continuous learning, accountability, and reliability.

24

1 Comment

New from Packt Publishing >> "Building Neo4j-Powered Applications with LLMs: Create LLM-driven search and recommendations applications with Haystack, LangChain4j, and Spring AI" by Ravindranatha Anthapu and Siddhant Agarwal Available at https://amzn.to/4l9lLKO This LLM book is for database developers and data scientists who want to leverage knowledge graphs with Neo4j and its vector search capabilities to build intelligent search and recommendation systems. Working knowledge of Python and Java is essential to follow along. Familiarity with Neo4j, the Cypher query language, and fundamental concepts of databases will come in handy. 𝓦𝓱𝓪𝓽 𝓨𝓸𝓾 𝓦𝓲𝓵𝓵 𝓛𝓮𝓪𝓻𝓷: 🟠Design, populate, and integrate a Neo4j knowledge graph with RAG 🔵Model data for knowledge graphs 🟢Integrate AI-powered search to enhance knowledge exploration 🟠Maintain and monitor your AI search application with Haystack 🔵Use LangChain4j and Spring AI for recommendations and personalization 🟢Seamlessly deploy your applications to Google Cloud Platform

42

3 Comments

I thoroughly enjoyed moderating a panel on quantum computing for executive leaders at SIBF this week in Palo Alto with Amit Bhattacharyya (UC Berkeley), Charles Chung (IBM), and Rajesh Ramanujam (Applied Ventures). We talked about a broad future "Beyond AI" and explore the various ways quantum is advancing. A few key takeaways: 1 - Chuck shared that IBM anticipates demonstrating "Quantum Advantage" next year, with a fault tolerant computer by 2029. 2 - Rajesh shared how companies concerned about data security should explore "Post Quantum Cryptography" 3 - Amit expounded upon how AI is accelerating quantum research, and that no matter how helpful AI may be today, students still must develop core critical thinking skills.

72

3 Comments

In the latest episode of the OpenAI podcast I got talk to OpenAI CFO Sarah Friar and legend himself Vinod Khosla about the AI industry. Sarah explained the relation between compute and revenue (OpenAI's revenue keeps 3x'ing each year) and Vinod provided an insightful way to tell if we're in a bubble: API calls. Are they going up or down? (Spoiler: Up, up, up!) https://lnkd.in/g2tTQBdp

23

1 Comment

The pace of capital flowing into AI feels familiar. These two charts from Apollo Academy highlight just how quickly expectations (and valuations) have risen. History tells us that periods of exuberance are almost always followed by a correction. At Buoyant, we believe both sides of this story can be true: ✅ Yes, the current surge of investment is likely ahead of itself. ✅ And yes, the underlying transformation is as inevitable as the Internet’s rise 25 years ago. We’ve seen this same pattern in climate tech: early hype, painful correction, and then enduring progress as the world reorganizes around a fundamental shift. The companies that endure are the ones that solve real problems with staying power. Our responsibility as investors at Buoyant Ventures is not to chase the froth, but to stay grounded in fundamentals — backing entrepreneurs who are building durable businesses that will still matter long after the cycle cools. AI’s trajectory may be uneven, but like climate tech, its long-term impact is unstoppable. 🤛 https://lnkd.in/gga2igQJ https://lnkd.in/gfmRy-VG

30

3 Comments

Why should Active Metadata Tags be the backbone of your AI strategy? Active Metadata Tags (AMTs) are used for automatic enrichment of metadata. Euno supports two types of custom metadata enrichment capabilities, both of which generate AMTs: #1 Extract structured metadata from unstructured metadata. This involved parsing metadata from elements like code and descriptions. A common application would be to enforce data modeling guardrails. Examples: → Implicit columns schema: Analyze dbt model code to determine whether the resulting schema includes implicit or explicit columns. → Verification checks: Assess naming conventions, detect use of approximate count distinct when querying large tables, evaluate documentation agility, flag full outer joins, and more. #2 Compose new AMTs. This approach combines metadata of a resource, and of any related resources, to set a new property on a resource. Examples: → Non-trivial custom SQL: Tag Tableau Custom SQL nodes that do not add logic. → Certified data source: Tag Tableau data sources as ‘certified’ if they don’t contain non-trivial custom SQL. → Certified dashboard: Tag Tableau dashboards as ‘certified’ if they only depend on certified data sources.   This capability is essential for governance workflows in large organizations where different teams maintain different parts of the stack. So team AI can build on something that team B created. Consider this scenario: → The team maintaining dbt can create a custom property to flag models for deprecation (e.g., models lacking both ‘prod’ and ‘prod_ready’ tags). → Analytics teams can then use this signal to create custom properties for their BI assets to drive workflows to migrate away from those models. Why does this also matter for AI agents? By implementing AMTs to surface trust signals, AI agents can actually tell the difference between exploratory work and official definitions. Example: → Euno’s MCP server surfaces to AI agents only Tableau logic that meets your trust standard and is considered certified such as: ‘depends only on dbt models from the gold layer and was in use in the past 60 days.’ *** Expect to hear much more about Active Metadata Tags! They’re quickly becoming the key to scale AI beyond the sandbox in your organization. #ActiveMetadata

17

1 Comment

Bain: Skyrocketing AI demand creates unattainable CapEx requirement at current compute cost; major technological innovation is needed to reduce cost of compute to keep up. (Spoiler: Investing in some of this at Jumpspeed Ventures ...) https://lnkd.in/dcS8Xddy

12

UC Berkeley just published the 1st large-scale study of agents in production, with input from 306 practitioners from 26 industries (mostly Banking and Tech). Agents are being kept deliberately simple because simple is debuggable. Reliability is the top challenge, requiring close supervision from a human-in-the-loop: 68% execute at most 10 steps before requiring human intervention 70% rely on prompting off-the-shelf models 74% depend primarily on human evaluation Humans are a more important part of these systems than many realize - it’s why we added annotation queues to the HoneyHive platform. Teams are throttling autonomy because they can't verify correctness at scale. Your agent’s ceiling is set by your ability to observe, evaluate, and debug.

20

2 Comments

The next frontier of AI isn't just about choosing the right model. It's about building moats through implementation. Across the Emergence Capital portfolio, companies like Bland and Federato are scaling faster by embedding Forward-Deployed Engineers. These engineers bridge the gap between AI and reality, ensuring real results in production. Read more from my colleagues Gordon and Kyle on why FDEs are the key to unlocking AI's full value. https://lnkd.in/gcsRRGX2

29

3 Comments

We live in an era of manufactured growth. OpenAI now sits at the center of a fragile AI economy built around circular financing and valuation distortion. The last time we saw something similar was in 2000 that led to the collapse of Lucent and Cisco Background: In 1997, Lucent and Cisco discovered a clever trick to sell more routers and switches during the internet boom They began lending to their own customers (vendor financing) This resulted in massive short term growth, until the bubble burst When customers defaulted Lucent and Cisco were left with bad debt. As a result - Lucent’s stock fell 99% and CIsco’s 80% What was once being cited as business innovation was suddenly seen as a ponzi scheme Fast forward to 2025. OpenAI playbook looks eerily similar Hidden Revenue Loops that nobody is talking about : 1. Microsoft <> OpenAI : Microsoft owns 30% of OpenAI and almost half of OpenAI’s costs are on Microsoft Azure. Every invested dollar cycles back as revenues 2. Anthropic <> AWS : Amazon invested $4B into Anthropic and Anthropic committed to AWS as it’s primary cloud provider. Anthropic’s spending boosts AWS revenues - a closed loop growth flywheel 3. NVidia <> OpenAI : Nvidia pledges $100B to OpenAI so OpenAI can build 10 GW of data centers. Every $10B invested by Nvidia generates $35B in GPU sales - another self financed growth loop masqueraded as organic demand 4. AMD <> OpenAI : OpenAI agreed to buying AMD chips and received warrants equal to 10% in AMD shares. As AMD’s stocks rise due to new revenues from OpenAI, OpenAI’s paper value inflates - valuation distortion in action 5. Oracle <> OpenAI : OpenAI signs a $300B cloud deal with Oracle. Oracle buys Nvidia chips to power those data centers. Nvidia invests back into OpenAI. Capital -> Chips -> Cloud -> Capital. The manufactured growth loop completes Demand Domino Effect: If any one node falters, the illusion of infinite demand will collapse OpenAI is positioned to be the "epicenter" of any future AI crisis If AI bubble bursts and the demand for token declines, - OpenAI & Anthropic will face down rounds - Nvidia and AMD face inventory gluts - AWS, Azure, Oracle end up with under utilized infrastructure Takeaways: Our systems and institutions rarely handle checks and balances well Every generation comes up with their own language to justify excess and hype In 1999: It was the eyeballs and internet adoption In 2008, it was credit innovation and sub prime lending In 2026, it will be compute, AI and infrastructure spending    When capital moves in controlled loops, it creates the illusion of infinite growth and perpetual demand Until reality kicks in, leading to a crash However, It’s important to remember that crashes don’t kill innovation; they expose our limitations The dot-com bust didn’t end the internet; it made it stronger If and when the AI crash happens, it will not end AI Instead, it will give rise to some of the best AI companies of the future

117

36 Comments