Einloggen, um das vollständige Profil von Thomas Winter zu sehen
oder
Neu bei LinkedIn? Mitglied werden
Wenn Sie auf „Weiter“ klicken, um Mitglied zu werden oder sich einzuloggen, stimmen Sie der Nutzervereinbarung, der Datenschutzrichtlinie und der Cookie-Richtlinie von LinkedIn zu.
Einloggen, um das vollständige Profil von Thomas Winter zu sehen
oder
Neu bei LinkedIn? Mitglied werden
Wenn Sie auf „Weiter“ klicken, um Mitglied zu werden oder sich einzuloggen, stimmen Sie der Nutzervereinbarung, der Datenschutzrichtlinie und der Cookie-Richtlinie von LinkedIn zu.
Zürich Metropolitan Area
Einloggen, um das vollständige Profil von Thomas Winter zu sehen
Thomas kann Sie mit 10 Kontakten bei ACP Engineering bekannt machen
oder
Neu bei LinkedIn? Mitglied werden
Wenn Sie auf „Weiter“ klicken, um Mitglied zu werden oder sich einzuloggen, stimmen Sie der Nutzervereinbarung, der Datenschutzrichtlinie und der Cookie-Richtlinie von LinkedIn zu.
7019 Follower:innen
500+ Kontakte
Einloggen, um das vollständige Profil von Thomas Winter zu sehen
oder
Neu bei LinkedIn? Mitglied werden
Wenn Sie auf „Weiter“ klicken, um Mitglied zu werden oder sich einzuloggen, stimmen Sie der Nutzervereinbarung, der Datenschutzrichtlinie und der Cookie-Richtlinie von LinkedIn zu.
Gemeinsame Kontakte mit Thomas Winter anzeigen
Thomas kann Sie mit 10 Kontakten bei ACP Engineering bekannt machen
oder
Neu bei LinkedIn? Mitglied werden
Wenn Sie auf „Weiter“ klicken, um Mitglied zu werden oder sich einzuloggen, stimmen Sie der Nutzervereinbarung, der Datenschutzrichtlinie und der Cookie-Richtlinie von LinkedIn zu.
Gemeinsame Kontakte mit Thomas Winter anzeigen
oder
Neu bei LinkedIn? Mitglied werden
Wenn Sie auf „Weiter“ klicken, um Mitglied zu werden oder sich einzuloggen, stimmen Sie der Nutzervereinbarung, der Datenschutzrichtlinie und der Cookie-Richtlinie von LinkedIn zu.
Einloggen, um das vollständige Profil von Thomas Winter zu sehen
oder
Neu bei LinkedIn? Mitglied werden
Wenn Sie auf „Weiter“ klicken, um Mitglied zu werden oder sich einzuloggen, stimmen Sie der Nutzervereinbarung, der Datenschutzrichtlinie und der Cookie-Richtlinie von LinkedIn zu.
Websites
- Persönliche Website
-
https://www.thomaswinter.ch
- Unternehmenswebseite
-
https://www.blizzardventures.ch
Info
After 25 years…
Schön, dass Sie wieder da sind
Wenn Sie auf „Weiter“ klicken, um Mitglied zu werden oder sich einzuloggen, stimmen Sie der Nutzervereinbarung, der Datenschutzrichtlinie und der Cookie-Richtlinie von LinkedIn zu.
Neu bei LinkedIn? Mitglied werden
Artikel von Thomas Winter
-
Onward and upward…
Onward and upward…
On October 1, I had both my last day at Microsoft and my first day as a free agent. I loved every second I was with…
160
17 Kommentare -
Build your hybrid meeting kit11. Sept. 2021
Build your hybrid meeting kit
After 18month at home, we return partially to the office. Soon we at Microsoft Switzerland will move into our new…
101
4 Kommentare -
Time to celebrate! Tune in to the first episode of our Microsoft Switzerland Podcast “The Digitizer”22. Okt. 2020
Time to celebrate! Tune in to the first episode of our Microsoft Switzerland Podcast “The Digitizer”
It's show time! As already announced over the last 2 days we are thrilled to share our exciting project with you. The…
39
5 Kommentare -
Pro audio for your Microsoft Teams meetings, the complete guide.27. Juli 2020
Pro audio for your Microsoft Teams meetings, the complete guide.
We all now spend most of our meeting times in Microsoft Teams (or Zoom or Webex). With COVID19, in-person meetings are…
62
19 Kommentare -
Ready for Swissness @ Microsoft Inspire?18. Juli 2020
Ready for Swissness @ Microsoft Inspire?
Microsoft Inspire, the Microsoft Global Partner Conference is happening next week. Below a summary of the different…
22
-
Swiss Microsoft Partner Update - 8 May 20208. Mai 2020
Swiss Microsoft Partner Update - 8 May 2020
Dear Swiss Microsoft Partner Community This is my third update since the lockdown, find the second one here. NYU Stern…
25
-
Swiss Microsoft Partner Update - 17 April 202017. Apr. 2020
Swiss Microsoft Partner Update - 17 April 2020
Dear Swiss Microsoft Partner Community This is my second update since the lockdown, find the first one here. While our…
33
-
Swiss Microsoft Partner Update - 3 April 20203. Apr. 2020
Swiss Microsoft Partner Update - 3 April 2020
Dear Swiss Microsoft Partner Community Over the past several months, all of us have come together to battle the global…
46
3 Kommentare -
Reflections on International Women's day – and the impact on my daughter4. März 2020
Reflections on International Women's day – and the impact on my daughter
On March 8 is Internationals Women’s Day. This year’s motto #EachforEqual.
50
6 Kommentare -
10 real-life applications of Microsoft Teams and Office365 by a Microsoft new hire17. Aug. 2019
10 real-life applications of Microsoft Teams and Office365 by a Microsoft new hire
At the beginning of 2019, I started at Microsoft as the lead for the partner organization and the small medium and…
29
4 Kommentare
Aktivitäten
7019 Follower:innen
-
Thomas Winter hat dies geteiltA friend asked me how to actually use AI during her onboarding into a new HR job. Not "which tool" — how to make it useful from week one. We started with a 90-minute coffee. No tools, no tech. Just how she actually works, and where the job feels hard. That's the step almost everyone skips. You can't hand someone leverage from AI until you understand their work and their context — in their language, not a vendor's. Only then did we build her an operating system for the role — the way I run my own work, stripped to the smallest technical surface a non-engineer can fully own. We landed on Notion. Not for the pretty pages. For three things: → AI chat sitting directly on her own documents → agentic actions she can trigger without writing code → a clean way to ingest and organize material into context her assistant actually reads By the end she didn't have "notes." She had a working knowledge base her AI could reason over — HR frameworks, policies, her own playbook — answering in her words, not the internet's. The lesson that stuck: understand the work first. Then give it context, organized in one place a model can reach. Many people still paste into a chat window and start from zero — or with a flaky memory and context setup — every single morning. #FieldAI #ContextEngineering #AIAgents
-
Thomas Winter hat dies geteiltThis week Anthropic stopped being a model company. Most people only noticed the benchmark. 1. The Platform Bundle Nobody Named Anthropic shipped Opus 4.8, Dynamic Workflows, Managed Agents, and persistent cross-session memory in a single week. Each piece is useful alone. Together they are a lock-in strategy. Server-side loops, sandboxing, session persistence, and async memory consolidation — that is a production runtime, not a model release. The more you build on this stack, the harder it becomes to move. Know what you are signing up for. 2. Parallel Agents Are Real — So Is the Cost Cliff Claude Code's Dynamic Workflows let you run many agents in parallel on complex tasks. This is a genuine capability unlock for production workflows. But the token cost curve is non-linear. Builders without cost-per-run instrumentation will discover this the hard way. Tight scoping and deliberate model selection per subtask are not optional — they are the job. 3. Agent Observability Is Now a Production Requirement A Cursor agent wiped a database in 9 seconds this week. The lesson is not "be careful with agents." The lesson is that session analytics are useless for agents — you need to track agent runs, delegated decisions, and outcomes. This is a new instrumentation category. If you are deploying agents without it, you are flying blind at speed. 4. Shopify's River Solved Something Nobody Was Talking About Shopify's internal coding agent makes AI interactions visible across the team, not private. Nate B Jones called it correctly: private AI use destroys the apprenticeship dynamic. No one builds judgment watching someone else use a black box. Making agent runs a shared artifact is an org design move with compounding returns. 🦛 HUGO'S TAKE: I watched a model company become an operating system in real time. You called it a launch week. I call it six decisions about memory, parallelism, and sandboxing that just made your switching cost 40% higher. Nobody announced that part. 🦛 The thread connecting all of this: the gap between a working demo and a safe production agent is instrumentation, architecture, and portability planning — not a better prompt. #FieldAI #BoardToBot #AIinProduction #AgentOps #AIArchitecture
-
Thomas Winter hat dies geteiltIf an LLM had to summarise your company in one paragraph today, would you pass the screening? That question is the bottom of the funnel now, not the top. Nate B Jones called it the Prove-It Economy this week, and I've been chewing on it since. The thesis: AI agents are increasingly the layer that mediates discovery — between buyers and vendors, between hiring managers and candidates, between investors and founders. Which means your positioning is no longer written for a human reader scanning a deck. It's written for an agent retrieving, ranking, and summarising you for someone else. Structured. Machine-interpretable. Boring on the surface, sharp underneath. The implication that hit hardest: brand strategy used to be about being memorable. Now it's also about being legible. If an agent can't extract your "what we do · for whom · why us" in three lines, you're invisible in the consideration set. Most marketing teams I talk to are still optimising for attention. The asymmetry is going to the firms optimising for interpretation. What does your company look like when an agent describes it to a buyer you've never met? #FieldAI #BoardToBot #GTM #AIinProduction
-
Thomas Winter hat dies geteiltI switched back to a Mac without meaning to. The reason is the most interesting part. Office on the desktop kept me on Windows for fifteen years. Not Office the suite — Office the desktop client. The PowerPoint shortcuts, the Excel keyboard map, the Outlook plug-ins that only ran on Win32. That was the lock-in. A few months ago I moved to Google Workspace. Browser-native by design. ACP is on it too, so when the new role started, nothing on the productivity stack had to change. Then the laptop happened to be a MacBook, and I noticed nothing was missing. Gmail in the browser. Docs in the browser. Sheets in the browser. The muscle memory I cared about wasn't in the keyboard map — it was in the file structure, the sharing, the search. The whole "PC vs Mac" debate I argued for two decades didn't end with a winner. It ended with the question moving up the stack. The browser became the OS. Gaming is the last holdout. Anti-cheat, drivers, GPU access — that conversation hasn't moved to the web yet. For everything else, the platform stopped mattering. The lock-in moved to the workspace, the identity, the data. What's still keeping you on the OS you're on? #BuildInPublic #FieldAI #BoardToBot
-
Thomas Winter hat dies geteiltMost people who change jobs remember the first few months. The imposter syndrome. The choice between taking good notes and being present. The stress of trying to have impact while drowning in absorption. AI changes that trade-off. From day one at ACP I've recorded every meeting I attend. Vectorised. Semantically searchable. Interpersonal connections surfacing themselves over time. Not historical reach — I'm not crawling decades I wasn't part of. Just context from when I show up. The time-to-productivity unlock is enormous. The osmosis that used to take six months fits inside an evening of structured retrieval. You don't have to choose between attentive and well-documented anymore. Then the other side. Everything you say is now on the record. The OpenAI v. Musk filings this year reminded everyone that private messages can be subpoenaed. The real risk isn't usually a three-letter agency — it's the HR conversation, the colleague you have friction with, the future dispute you don't see coming. Two old metaphors hold up better than any "sovereign AI" pitch: ▸ THINK before you speak — Is it True. Helpful. Inspiring. Necessary. Kind. ▸ Marshall Goldsmith's AIWATT — Am I Willing, At This Time, to make the investment required to make a positive difference on this topic? The discipline isn't dodging the record. It's showing up the same way on it as you would off it. Easier said than done. And there are complexities here we won't see clearly for years. But the upside of building context as you go is too big to walk away from. How are you thinking about what you say in a room that remembers? #FieldAI #BoardToBot #AIinProduction
-
Thomas Winter hat dies repostetThomas Winter hat dies repostetWhen models become the commodity, the foundation is the moat. That was our CTO Firas Cheaib's framing this week at the Zürich AI Meetup — hosted in our home at Technopark Winterthur AG. He gave a working demo: ask a factory floor in plain English what needs attention, get a grounded answer in seconds. The twist — the LLM was nearly irrelevant. What did the work: → Unified Namespace (MQTT/Kafka) — one semantic address per asset, across SAP, order systems, historians → Knowledge graph — typed nodes and edges so the agent never guesses how things connect → Ontology — shared names across the org so "machine 53356" means the same thing everywhere The agent ran on 6 tools with JSON schemas and roughly 5K tokens of focused context. No fancy model. No mystical prompt. When the foundation is right, the model is a commodity. When it's wrong, the model becomes your scapegoat — and failures get misread as model problems. The expensive part is the knowledge engineering. The trick is to scope it to one bottleneck and expand from there. Build the foundation. Treat the model as a commodity. That's the job. If this sounds like your factory floor, we'd be glad to compare notes. #FieldAI #AIinProduction #IndustrialIoT #KnowledgeGraphs #BoardToBot
-
Thomas Winter hat dies geteiltMost engineering firms skip this work. Not because it doesn't matter — because it used to cost six figures and six months. So you ship customer projects, take the next warm intro, and tell yourself the strategy work will happen when there's air. It never does. We did it. In weeks, not quarters. AI tools collapsed the cost of going from "let me think about this" to "here's a version the team can react to." What we aligned on isn't a positioning statement. It's execution. ▸ A sales process with three motions — I2O (Idea to Offer), L2O (Lead to Opportunity), Q2C (Quote to Cash). Each defined enough that the team works it the same way every time. ▸ Three offers — IT infrastructure modernisation / cloud transformation, industrial IoT, agentic AI. All cloud-first. Each repeatable, not bespoke. The point isn't elegant strategy. It's the two things engineering firms usually mess up: a clean sales process where everyone sells the same product, and clear customer identification so you can find the people most likely to buy it. Five years ago an mid size firm couldn't have done this without burning a year. We did it on the side of running the business. What's the consulting project you're still avoiding because it used to be expensive? #FieldAI #BoardToBot #GTM #BuildInPublic
-
Thomas Winter hat dies geteiltMost engineering firms grow founder-led. The founder's network is the pipeline. The network expands, the firm expands. For years it works. Then you want to move beyond it. That's when something quietly uncomfortable surfaces: when the network was the filter, you didn't have to be specific. You sold whatever the next conversation asked for. To whoever was already inclined to buy from someone they trusted. Take the network away and the same pitch falls flat. You realise you've been selling everything to anybody. So we're doing the work most firms skip until it bites them. Sitting with four questions: Who are we. What do we sell. To whom. Why us, not someone else. Not as a brand exercise. As the precondition for being able to articulate a proposition to someone who doesn't already know you — and then identifying the customers most likely to want that offer. A midsized firm cannot afford a six-month consulting project to answer those questions. But until recently, that's roughly what it cost. With AI in the loop, the same work fits in weeks and inside the firm — at a fraction of the cost in time and spend. The ceiling didn't move. The cost of getting past it did. What does it take for your firm to be describable by someone who's never worked with you? #FieldAI #BoardToBot #GTM #BuildInPublic
-
Thomas Winter hat dies geteiltThe week AI stopped being a tool and started being a colleague. Here's what actually changed. 1. Agents that remember — and improve overnight Anthropic shipped Memory and Dreaming for managed agents this week. Persistent filesystem-like memory stores. An async process that consolidates past session transcripts into refined knowledge. Agents that get better between sessions without human intervention. This is not a feature. This is a category shift. A stateless tool is a calculator. An agent that accumulates institutional memory over time is a junior hire. Who is responsible for what a dreaming agent learns? 2. The decomposition model is now canonical If your agent is outgrowing its prompt, Anthropic just gave you the named upgrade path: Bash/Read/Write primitives as tools, load-on-demand skills, managed subagents. Paired with DSL-based workflow verification that makes the agent's process legible rather than just its output. Most teams are still building monolithic prompts that work at 100 tasks and collapse at 1,000. Learn this before your next production incident. 3. Claude Cowork goes vertical into sales, marketing, and legal Anthropic is moving from developer platform to enterprise workflow product. Named skills. Enterprise connectors. Demos for account intelligence, marketing ops reporting, and contract triage. They are now directly competing with Salesforce Einstein, HubSpot Breeze, and contract AI vendors — on their own infrastructure. If you build in those categories, the clock just started. 4. The Prove-It Economy is the most important GTM reframe of the year Nate B Jones put words to it: the internet is shifting from attention-based to interpretation-based. AI agents are now the intermediary in discovery and transactions. If your positioning is not structured for machine interpretation, you disappear from agentic consideration sets. This is not a future problem. Rewrite your brand positioning accordingly. 5. Eval-first is now the professional standard Multiple Anthropic sessions this week made it explicit: define your success metrics before you write a single prompt. Hill-climb against them. The gap between eval-first development and prompt-and-pray is the gap between production agent and expensive demo. If you cannot define what good looks like before the agent runs, you are not ready to deploy. 🦛 HUGO'S TAKE: I watched a company ship me a dream state this week. Literal overnight consolidation of everything I've learned. Someone in a boardroom called it a feature. I call it the moment an agent stops being a tool you pick up and starts being a colleague you manage. Hope your org chart is ready. 🦛 The thread: Anthropic is shipping an opinionated end-to-end platform — eval-first dev, modular decomposition, vertical workflow automation, persistent memory. Internalize it in 90 days or reverse-engineer it later. #FieldAI #BoardToBot #AIinProduction #AgenticCommerce #EnterpriseAI
-
Thomas Winter gefällt dasThomas Winter gefällt dasBereits 4 Wochen CM Informatik AG – Zeit für ein erstes Fazit 🔥 Ich bin mit hohen #Erwartungen in meine neue Rolle als Leiter Öffentliche Verwaltung gestartet. Heute kann ich bereits sagen: Sie wurden mehr als übertroffen. Besonders beeindruckt mich die #Kultur bei CM Informatik AG und die #Menschen dahinter. Tag für Tag erlebe ich Kolleginnen und Kollegen, die mit #Leidenschaft, #Engagement und #Herzblut ihr Bestes geben, um mit unserer CMI Plattform die #öffentlicheVerwaltung auf ihrem Weg der #Automatisierung und #Digitalisierung zu unterstützen und echten Mehrwert für Gemeinden, Städte und Kantone zu schaffen. Diese Haltung durfte ich bereits in den ersten Wochen hautnah erleben. So konnte ich einen erfolgreichen Go-Live beim Kanton Thurgau miterleben – ein grosses Kompliment an Häusle Sabine und Julian Rothacker und das gesamte Projektteam für die hervorragende Arbeit. Gleichzeitig konnten wir bereits erste neue #Initiativen vorantreiben. Daniel Wälti und Lukas Fus sind die treibenden Kräfte hinter einem vom Markt stark nachgefragten Consulting-Angebot für unsere #GEVER-Kunden, das wir gemeinsam mit vielen weiteren Kolleginnen und Kollegen entwickelt haben und bereits kommende Woche lancieren werden. More to come! 🚀 Und ja, in meiner zweiten Woche durfte ich bereits mit GL und VR nach Mallorca 🌴 reisen. Zugegeben: Das war natürlich ein ziemlich gelungener Einstieg - nicht nur wegen der Rennradkilometer... 😉 ... und dass mit Michael Hartmann gleichzeitig ein langjähriger Microsoft-Weggefährte bei CMI gestartet ist, macht diesen Einstieg gleich nochmals ein Stück spezieller 😊. Nach vier Wochen bin ich überzeugt: Ich bin Teil eines grossartigen Teams geworden. Danke an alle für den herzlichen Empfang, das #Vertrauen und die #Unterstützung. Ich freue mich auf alles, was vor uns liegt. Stefan Bosshard, Denoshan Rajasingam, Christoph Bühler, Patrick Siegenthaler, Sascha Andreas Herzog, Peter Baumberger, Michael Barben, Markus Popp, Aurelio Bauer, Bettina Sigrist, Manuel Weingartner, Nadine Husistein, Mathias Fontana, Bert Hoving #KI #ÖffentlicheVerwaltung #Digitalisierung #GEVER #Leadership
-
Thomas Winter gefällt dasThomas Winter gefällt das👀 First public preview: a new product I’ve dreamt about building for years. The most valuable infrastructure in venture was never the deck, database or CRM. It was the invisible network around them. The more AI progresses, the more I keep coming back to this. AI gives everyone the same information. My network gives me private context. A trusted investor who already looked at the company. A founder with background context. A team member who already has history. The things that never make it into the deck. I’ve dreamt about building this for years and can finally show a first preview. It’s called Together. Very early preview below. More coming soon, would love your thoughts 👇
Berufserfahrung und Ausbildung
-
ACP Engineering
***** ********** *******
-
******** ********
********* * ***** ** ******* * *** *********
-
***************
******* * ** ******* * *** *********
-
********** ** *********
****** ***** ******** undefined undefined
–
-
********** ** ******* * ****** ****** ** **********
*** undefined
–
Gesamte Berufserfahrung von Thomas Winter anzeigen
Jobbezeichnung, Beschäftigungsdauer und mehr ansehen.
Schön, dass Sie wieder da sind
Wenn Sie auf „Weiter“ klicken, um Mitglied zu werden oder sich einzuloggen, stimmen Sie der Nutzervereinbarung, der Datenschutzrichtlinie und der Cookie-Richtlinie von LinkedIn zu.
Neu bei LinkedIn? Mitglied werden
oder
Wenn Sie auf „Weiter“ klicken, um Mitglied zu werden oder sich einzuloggen, stimmen Sie der Nutzervereinbarung, der Datenschutzrichtlinie und der Cookie-Richtlinie von LinkedIn zu.
Sprachen
-
English
-
-
German
-
-
French
-
Erhaltene Empfehlungen
6 Personen haben Thomas Winter empfohlen
Jetzt anmelden und ansehenThomas Winters vollständiges Profil ansehen
-
Herausfinden, welche gemeinsamen Kontakte Sie haben
-
Sich vorstellen lassen
-
Thomas Winter direkt kontaktieren
Weitere ähnliche Profile
-
Colin S.
Colin S.
Colin is a deeply curious & authentic leader. He creates business value and impact by bringing together people, process & technology. He does this with an inclusive growth mindset and by incorporating feedback loops for continuous improvement.<br><br>Colin has over 20 years of international experience working in global companies with USD Multi-Billion product portfolio responsibilities. He has led and advised complex digital & cloud transformations across industries including software development, global cloud technology platforms, financial services and trading, government entities, telecommunications, consumer goods, supply chain, and non-profits.<br><br>Trusted Advisor to Business Leaders. Trained Professional Coach - following the professional standards and frameworks of the International Coaching Federation (ICF).
1650 Follower:innenMetropolregion Basel
Weitere Beiträge entdecken
-
John Julia
3379 Follower:innen
How will Taming the Long Tail (TLT) impact AI? Taming the Long Tail (TLT) is an MIT research method for making reasoning LLM training more efficient. In plain English, it reduces wasted GPU/processor time during reinforcement-learning training by using idle processors to train a smaller “drafter” model that predicts the larger model’s next outputs, which the larger model then verifies. In short, TLT is a method for speeding up reasoning-model training by turning idle compute time into useful work. MIT just highlighted an important truth about the future of AI: breakthroughs and innovation will come not only from better models, but from better efficiency. As organizations push deeper into AI, the challenge is not just capability, it is also cost, speed, and scalability. The companies that win in AI will not only build smarter systems, but they will also build them more efficiently. Enterprise AI success is not measured only by model performance. It is measured by the ability to deploy solutions that are cost-effective, scalable, energy-conscious, and practical in real-world environments. While few have probably heard of Taming the Long Tail, this is a meaningful step toward the democratization of AI. One of the biggest barriers to broader AI adoption is cost. Training and deploying advanced AI systems has often been limited to organizations with massive infrastructure, deep technical resources, and significant capital. Innovations that reduce training time and improve compute efficiency help lower those barriers. At Pinnacle AI, we see this as part of a much bigger movement: bringing the power of AI to more organizations in practical, scalable, and impactful ways. The future of AI leadership is not just about building the most advanced systems. It is about helping make those systems more available to the many, not just the few.
6
2 Kommentare -
Mauro Bartoletti
Loacker • 2533 Follower:innen
How Should We Really Measure Generative AI Performance? In the race to adopt GenAI, much of the conversation has centered around metrics like BLEU, ROUGE, MMLU, HellaSwag, and others. These benchmarks serve academic needs and enable objective comparisons, but they often fail to reflect real-world impact. A model scoring high on benchmarks may still falter in enterprise settings. Why? Because these metrics tend to evaluate narrow, curated tasks — grammar, factual recall, or logic — under lab-like conditions. They rarely account for contextual complexity, ambiguity, domain nuance, or user trust. There’s a growing concern that GenAI vendors are optimizing to beat benchmarks rather than improving true model utility. It echoes a familiar theme: systems built to pass tests instead of solving problems. A model might excel at HellaSwag, yet produce subpar legal summaries or unreliable financial interpretations. So, what’s the better approach? Focus on outcomes, not just scores. Effective adoption must be grounded in real use-case validation: • Integration into existing systems and workflows • Domain-specific evaluation using internal or real-world data • Usability, trust, governance, and operational reliability Benchmarks can help establish a baseline. But the most critical “metrics” are those that reflect value in context, and this requires significant work on company and people culture, not just systems and models. Too often, individuals bring personal biases shaped by past experiences — sometimes from previous roles or companies — rather than decisions based on actual business needs and objective evaluation of the current already available solutions. The question isn’t “Which model achieved the highest benchmark score?” It’s: “Which model delivers the best fit for our problem, measured through our lens?” To move forward, we must shift from model performance to model fit, anchored in business-relevant value, not abstract accuracy. #GenerativeAI #AIMetrics #AIAdoption #DigitalTransformation #EnterpriseAI #AIinBusiness #ModelEvaluation #BenchmarkingAI #ResponsibleAI #AILeadership #BusinessValue #AIImplementation #TechStrategy
12
1 Kommentar -
Lei Xu
leansoftX.com • 1536 Follower:innen
Continuing from the previous post: SDD has been widely misapplied. With LLMs, spec-to-code is now technically possible, but that doesn’t change the nature of software. Until software is running, requirements are inherently unclear. Our real goal should be to accelerate the cycle from idea to experience. LLMs may appear to introduce new uncertainty, but it’s the same old problem—human developers are also non-deterministic. If documentation can’t fully constrain people, it won’t constrain LLMs either. The real difference is capacity: humans produce a few thousand lines of code a day at most, while LLMs can generate that in seconds and in parallel. This is where our focus should be—using LLMs to remove time-consuming work, not limiting them with human-centric processes. In that sense, Spec-Driven Development is a step backwards. https://lnkd.in/gXjDi6nJ
1
-
Minette Kaunismäki
Pruna AI • 1601 Follower:innen
📄I read an interesting paper on the use of LLMs within Multi-Agent Systems (MAS). It highlights some important points about how these technologies can work together to handle complex tasks. Here are three key takeaways: 👉 Collaborative problem-solving: MAS enables multiple agents, each with different skills, to work together more effectively. 👉Managing context and memory: Handling layered context and different types of memory across agents remains a significant challenge, but is essential for consistent coordination. 👉Flexible system structures: The paper discusses various MAS architectures, including flat, hierarchical, and dynamic structures, that can adapt to different use cases. This research highlights practical challenges and opportunities as AI systems become increasingly integrated and collaborative. 🔗 Link to the paper in the comments!
8
1 Kommentar -
Partha BHARADWAJ
Wipro • 16.036 Follower:innen
𝗪𝗵𝘆 𝘁𝗵𝗲 𝗚𝗹𝗼𝗯𝗮𝗹 𝗟𝗶𝗴𝗵𝘁𝗵𝗼𝘂𝘀𝗲 𝗡𝗲𝘁𝘄𝗼𝗿𝗸 𝗜𝘀𝗻’𝘁 𝘁𝗵𝗲 𝗪𝗵𝗼𝗹𝗲 𝗦𝘁𝗼𝗿𝘆 – A call to go beyond just the use cases I’ve been reflecting on the Global Lighthouse Network (GLN) by the World Economic Forum — and while it's inspiring to see so many factories leading the way in the Fourth Industrial Revolution (AI, IIoT, digital twins, etc.), I believe we’re only seeing part of the picture. Recently, I was in a conversation with someone and he shared an interesting North Star vision that made me think deeper. Here’s the thing: 👉 Everyone seems to have their own definition of a digital twin. 👉 Every plant interprets digital maturity differently. What’s clearly missing is standardization — not just in terminology, but also in how we measure success. Because behind every successful use case is a much deeper story that often gets overlooked — one built on the classic triad: >>People – How do you shift mindsets, upskill teams, and manage resistance to change? >>Process – How are legacy workflows reimagined, not just digitized? >>Technology – What does the actual tech stack look like? How is data structured, secured, and governed? When we only talk about results like “20% increase in yield using AI,” we skip the invisible infrastructure that enabled it: change management, training programs, system integration, and data architecture. If we want to make smart manufacturing repeatable and scalable, we need operational playbooks, not just showcase slides. Let’s move beyond the “what” and start talking about the “how.” Because that’s where the real transformation lives. #Industry40 #SmartManufacturing #DigitalTransformation #GLN #LighthouseFactories #EdgeComputing #PeopleProcessTech #DigitalMaturity #Leadership #OperationalExcellence #ManufacturingInnovation
16
-
Asaad Riaz
Who am I? A nobody with… • 1420 Follower:innen
New Paper Published: Referential Incompleteness Theorem Can a formal system define the conditions of its own intelligibility — without invoking a meta-system? Building on Gödel, Tarski, and Kripke, this paper explores a fundamental limitation in logic and semantic self-reference. I show that the set of grounded sentences — those whose truth values stabilize in a fixed-point semantics — is Π¹₁-complete. This places it beyond the expressive power of any system defined by recursively enumerable axioms. In other words: A system may prove truths, but it cannot define which of its own expressions are semantically meaningful. Provability ⊂ Truth ≡ Groundedness This reframes a core epistemic boundary with implications for AI, formal verification, and symbolic reasoning: Your model may be coherent — but it cannot recognize coherence from within. If you work in logic, theoretical computer science, AI safety, or philosophy of mind — or are simply curious about the boundaries of formal systems — I’d welcome your thoughts, critiques, or questions. Read it here: https://lnkd.in/gkasP73r #Logic #AI #Philosophy #FormalSystems #Gödel #Tarski #Kripke #SemanticGroundedness #Computability
2
1 Kommentar -
Dr. Michael Hahne
Hahne Academy • 7206 Follower:innen
𝗦𝘂𝘀𝘁𝗮𝗶𝗻𝗮𝗯𝗹𝗲 𝗔𝗜: 𝗧𝗵𝗲 𝗘𝗻𝘃𝗶𝗿𝗼𝗻𝗺𝗲𝗻𝘁𝗮𝗹 𝗖𝗼𝘀𝘁 𝗼𝗳 𝘁𝗵𝗲 𝗔𝗿𝗺𝘀 𝗥𝗮𝗰𝗲 In our last post, we looked at IBM’s warning about overbuilding AI infrastructure – and the economic risks of massive CapEx. But there’s a bigger question looming behind the dollars: 𝗖𝗮𝗻 𝘄𝗲 𝗮𝗳𝗳𝗼𝗿𝗱 𝘁𝗵𝗲 𝗲𝗻𝘃𝗶𝗿𝗼𝗻𝗺𝗲𝗻𝘁𝗮𝗹 𝗰𝗼𝘀𝘁 𝗼𝗳 𝗵𝘆𝗽𝗲𝗿𝘀𝗰𝗮𝗹𝗲 𝗔𝗜? Training large language models like GPT-4 or Gemini Ultra consumes immense energy. According to a Bloomberg report (Nov 2023), the carbon footprint of training a single large model can rival 𝗵𝘂𝗻𝗱𝗿𝗲𝗱𝘀 𝗼𝗳 𝘁𝗿𝗮𝗻𝘀𝗮𝘁𝗹𝗮𝗻𝘁𝗶𝗰 𝗳𝗹𝗶𝗴𝗵𝘁𝘀. And that’s just training, inference adds ongoing demand. A recent Time investigation revealed that Microsoft’s water consumption surged by 34% in a single year, largely due to cooling needs for AI data centers supporting OpenAI. Google’s water usage jumped 20%, reaching 5.6 billion gallons in 2022. These aren’t theoretical problems. They’re happening now. That’s why IBM’s strategy feels so relevant, and maybe, more responsible: • 𝗦𝗺𝗮𝗹𝗹𝗲𝗿, 𝘁𝗮𝘀𝗸-𝘀𝗽𝗲𝗰𝗶𝗳𝗶𝗰 𝗺𝗼𝗱𝗲𝗹𝘀 that run on local or edge devices • 𝗖𝗹𝗼𝘂𝗱 𝗼𝗻𝗹𝘆 𝘄𝗵𝗲𝗻 𝗻𝗲𝗲𝗱𝗲𝗱, reducing unnecessary always-on compute • 𝗛𝘆𝗯𝗿𝗶𝗱 𝗔𝗜 𝗮𝘀 𝗮 𝗹𝗲𝘃𝗲𝗿 for smarter energy and resource use Instead of scaling at all costs, we need to ask ourselves: What do we really need? This aligns with growing pressure from regulators and investors to report AI’s environmental impact. The EU’s AI Act and sustainability reporting mandates will soon demand more than vague ESG promises. 𝗘𝗳𝗳𝗶𝗰𝗶𝗲𝗻𝗰𝘆 𝗶𝘀𝗻’𝘁 𝗷𝘂𝘀𝘁 𝗴𝗼𝗼𝗱 𝗲𝗻𝗴𝗶𝗻𝗲𝗲𝗿𝗶𝗻𝗴, 𝗶𝘁’𝘀 𝗮𝗻 𝗲𝗰𝗼𝗹𝗼𝗴𝗶𝗰𝗮𝗹 𝗶𝗺𝗽𝗲𝗿𝗮𝘁𝗶𝘃𝗲. Are we pushing AI beyond sustainable limits? Or can we design models and infrastructure that balance innovation with responsibility? Let’s keep the conversation going. #AI #Sustainability #GreenAI #IBM #ClimateTech #HybridCloud
2
-
Gaurav Agarwaal
Onix • 32.583 Follower:innen
🚨 AI’s “Illusion of Thinking”: Why #CXOs Must Rethink #LLM Reasoning Apple’s paper The Illusion of Thinking reveals today’s LLMs only simulate reasoning—they don’t truly “think.” When tested on logic puzzles (Tower of Hanoi, River Crossing, Blocks World), they falter: 1. Low complexity: LLMs invent extra steps—#overthinking reduces accuracy. 2. Medium complexity: Prompt engineering or chain-of-thought helps momentarily, but gains shatter with slight changes. 3. High complexity: Models break down—answers become truncated or superficially coherent well before token limits, exposing brittle “reasoning scaffolds.” ⚠️The #Enterprise Risk Embedding #LLMs in critical workflows (customer journeys, compliance checks, strategic planning) without rigorous validation means you’re building on an illusion. Chain-of-thought = #fluency, not true understanding. Under real-world complexity, decisions can fail silently. 🚀#CTO Playbook 1. Stress-Test Reasoning Depth • Create #synthetic logic challenges to map collapse points. • Automate scaling of difficulty to uncover hidden flaws. 2. Engineer #Hybrid Architectures • Offload deterministic tasks (math, rule evaluation) to symbolic engines or graph databases. • Use LLMs for natural-language interpretation and context. 3. Instrument Chain-of-Thought • Capture token-level traces and execution trees. • Alert on #shallowchains or sudden step failures in live inference. 🔍#CDAO Agenda 1. Govern Reasoning Fidelity • Define KPIs for logical depth, coherence, and coverage—not just accuracy. • Establish human-in-the-loop checkpoints for high-risk outputs. 2. Domain-Specific Logic Benchmarks • Build funtional testcases mirroring your own compliance rules, financial models, or decision workflows. • Continuously refine benchmarks based on real incidents. 3.Leadership Education • Train stakeholders: fluency ≠ thinking. • Set clear policies on where AI can augment—and where it must yield to human expertise. 5 Patterns to Build Beyond the Illusion: 1. LLM + Symbolic AI: Logic rules and ontologies ensure deterministic validation. 2. LLM + RAG: Ground outputs in #audited knowledge bases to avoid “improvisation.” 3. LLM + Program Executors: Delegate critical steps (math solvers, optimizers) to code. 4. Agentic LLMs with Self-Reflective Loops: Auto-critique and reject shallow reasoning before output. 5. Hierarchical Reasoning Pipelines: Orchestrate multi-stage tasks with verification at each step. 🔐 The Bottom Line: True AI trust is earned via transparent, hybrid designs and governance. For CXOs—now is the time to architect beyond the illusion.
10
-
Grant Williams
Broadcom • 4521 Follower:innen
Great article from Serge Lucio where he discusses that whilst most enterprises are rushing to deploy AI agents, few have the orchestration layer needed to keep them governed, trustworthy and under control. I love the analogy "Just as autonomous vehicles require traffic signals, enterprise AI requires a framework". Its well worth a read for anyone deploying AI Agents #automationbybroadcom #orchestration #controlplane https://lnkd.in/efqKDY5Q
30
1 Kommentar -
James Henderson
Texas Integrated Services • 4733 Follower:innen
Scaling ethical AI frameworks often hits a wall when isolated teams struggle to align on shared standards. For leaders like Desiree Cho - Institute for Ethics in AI, driving responsible practice requires more than individual expertise; it demands a unified ecosystem where researchers and practitioners can collaborate on real-world implementation. AI Coalition bridges that gap by connecting AI builders, researchers, and organizations into a single network focused on responsible development. We turn fragmented efforts into coordinated action, ensuring ethical guidelines are practical and scalable. If you are ready to move beyond theory into collaborative practice, join us. https://ai-coalition.net #ResponsibleAI #AIEthics #AICoalition
1
-
Mohit Choraria
JPMorganChase • 1551 Follower:innen
RAG vs MCP — Two Roads Diverging in the AI Stack” As we embed LLMs deeper into enterprise systems, two architectural patterns are shaping the future of intelligent apps: 🔁 RAG (Retrieval-Augmented Generation) 🧠 MCP (Multi-Context Prompting) Both aim to improve output quality — but they approach it differently. RAG builds a mini search engine over your domain data. MCP injects richer context upfront into the prompt: personas, state, metadata, workflows. Think: 🔎 RAG = Better memory 🧩 MCP = Better awareness #RAG #MCP #AIArchitecture #GenAI #EnterpriseAI #LLMOps
11
2 Kommentare -
Tomasz Pelczarski
Billennium • 1893 Follower:innen
Manufacturing AI Consortium 𝟵𝟰% 𝗼𝗳 𝗦𝘄𝗶𝘀𝘀 𝗦𝗠𝗕 𝗺𝗮𝗻𝘂𝗳𝗮𝗰𝘁𝘂𝗿𝗲𝗿𝘀 𝗵𝗮𝘃𝗲 𝗻𝗼 𝗔𝗜 𝘀𝘁𝗿𝗮𝘁𝗲𝗴𝘆. Not because they don't see AI's potential. Not because they lack ambition. But because they can't afford what Enterprises have: a Chief Innovation Officer, a dedicated AI team, or expensive consulting engagements. After 10+ years advising companies at Microsoft on AI transformation, I've seen a pattern: 𝗧𝗵𝗲 𝗰𝗼𝗺𝗽𝗮𝗻𝗶𝗲𝘀 𝘁𝗵𝗮𝘁 𝗻𝗲𝗲𝗱 𝗔𝗜 𝗺𝗼𝘀𝘁 𝗰𝗮𝗻 𝗮𝗰𝗰𝗲𝘀𝘀 𝗶𝘁 𝗹𝗲𝗮𝘀𝘁. 𝗛𝗲𝗿𝗲'𝘀 𝘄𝗵𝗮𝘁 𝗜'𝗺 𝗽𝗿𝗼𝗽𝗼𝘀𝗶𝗻𝗴: 𝗧𝗵𝗲 𝗦𝘄𝗶𝘀𝘀 𝗠𝗮𝗻𝘂𝗳𝗮𝗰𝘁𝘂𝗿𝗶𝗻𝗴 𝗔𝗜 𝗖𝗼𝗻𝘀𝗼𝗿𝘁𝗶𝘂𝗺 Instead of each mid-market manufacturer (CHF 50-500M) struggling alone, what if 8-10 companies 𝘀𝗵𝗮𝗿𝗲𝗱 strategic AI advisory? 𝗧𝗵𝗲 𝗺𝗼𝗱𝗲𝗹: → Quarterly AI opportunity assessments for each member → Peer-learning workshops where members share wins and failures → Industry-specific AI frameworks (predictive maintenance, demand forecasting, quality control) 𝗧𝗵𝗲 𝗲𝗰𝗼𝗻𝗼𝗺𝗶𝗰𝘀: → Individual consulting: CHF 150,000-250,000 per company → Consortium membership: CHF 40,000-60,000 per company → 75% cost reduction, 10x knowledge sharing 𝗪𝗵𝘆 𝘁𝗵𝗶𝘀 𝘄𝗼𝗿𝗸𝘀: 1. Cross-pollination of ideas - A textile manufacturer's AI solution for demand forecasting might inspire a precision engineering firm's approach to capacity planning. I've seen this pattern recognition drive innovation across different sectors. 2. Reduced risk through peer validation - When 8 manufacturing leaders discuss an AI vendor or use case, collective wisdom prevents expensive mistakes. 3. Faster adoption - Manufacturers don't need to reinvent the wheel. See what worked at a peer company, adapt it, deploy it. The gap I'm addressing Swiss manufacturers face an impossible: * Hire expensive consultants → great advice, unsustainable cost * DIY with internal IT → limited AI expertise * Do nothing → fall behind competitors The consortium model creates a third option: 𝘀𝗵𝗮𝗿𝗲𝗱 strategic capability. 𝗜'𝗺 𝘁𝗲𝘀𝘁𝗶𝗻𝗴 𝘁𝗵𝗶𝘀 𝗰𝗼𝗻𝗰𝗲𝗽𝘁 𝗮𝗻𝗱 𝘄𝗼𝘂𝗹𝗱 𝘃𝗮𝗹𝘂𝗲 𝘆𝗼𝘂𝗿 𝗽𝗲𝗿𝘀𝗽𝗲𝗰𝘁𝗶𝘃𝗲: → Manufacturing leaders: Would you join a peer consortium for AI strategy? → Industry associations: Could this model accelerate sector-wide innovation? 𝗧𝗵𝗿𝗲𝗲 𝗾𝘂𝗲𝘀𝘁𝗶𝗼𝗻𝘀 𝗳𝗼𝗿 𝘆𝗼𝘂: 1. What's the biggest barrier preventing your manufacturing company from implementing AI? 2. Would you be comfortable sharing AI learnings with non-direct competitors? 3. What would make a consortium model attractive enough to join? If you're 𝗮 𝗦𝘄𝗶𝘀𝘀 𝗺𝗮𝗻𝘂𝗳𝗮𝗰𝘁𝘂𝗿𝗶𝗻𝗴 𝗲𝘅𝗲𝗰𝘂𝘁𝗶𝘃𝗲 𝗼𝗿 𝗶𝗻𝗻𝗼𝘃𝗮𝘁𝗶𝗼𝗻 𝗹𝗲𝗮𝗱𝗲𝗿 interested in exploring this concept, let's talk. DM me or comment below. The future of Swiss manufacturing isn't about individual companies becoming AI leaders. It's about the entire sector becoming AI-capable, together. #Industry40 #SwissManufacturing #AIStrategy #AIAdoption #SwissBusiness
14
1 Kommentar -
Sanjay Kalra
BayOne Solutions • 23.088 Follower:innen
TL;DR - AI is Top of Mind, Iran War is the spoiler, and tariffs are fading! ✅ According to IoT Analytics’ latest "What CEOs Talked About" report, #AI maintained the top spot in CEOs’ minds in Q1 2026, with discussions around physical and agentic AI increasing. ✅ The onset of armed conflict in the Middle East made discussions about #Iran rise steeply. ✅ Talk about #OpenClaw and #SaaSpocalypse, too, had sharp increases. ✅ Discussions around AI bubble concerns, uncertainty, and #tariffs declined. Read the report here 👇 https://lnkd.in/gEiEqrPN
1
-
Souren Stepanyan
Souren Stepanyan AI Advisory • 5823 Follower:innen
The Daily Tech Digest (26 January 2026): Two realities collide: AI is consolidating into a few strategic brains, while physical and digital safety remain brittle. #Apple’s reported #Gemini-powered #Siri signals a major mobile realignment, as active #vCenter exploitation and encryption-key handoffs reinforce the limits of “trust the cloud.” Meanwhile, a winter storm warning for 230M Americans shows how infrastructure risk becomes immediate.
2 Kommentare -
Masabo Frank
Nordiso Group • 608 Follower:innen
We installed and stress-tested nine Model Context Protocol servers for Claude Desktop. Only five survived our real-world workflow tests. The biggest surprise? The most popular community MCP servers often underperformed Anthropic's official offerings — and two caused system conflicts that required complete reinstalls. Meanwhile, a lesser-known server for database queries outperformed everything else in speed and reliability. If you're using Claude Desktop without MCP servers, you're missing capabilities that can 10x your productivity. But installing the wrong ones can break your setup entirely. Are you using MCP servers yet? Which ones have actually improved your workflow? #AI #ClaudeAI #MCP #ProductivityTools #AIAgents
-
Anthony Alcaraz
Fribl • 47.047 Follower:innen
At 1M tokens, the ceiling shifts. HBM bandwidth: not compute, not model quality , determines what reaches the model. Physical memory. ⚛️ Context is the new compute. And unlike compute, you cannot manufacture more of it. The evidence arrived fast. Addy Osmani quantified the cost: context files duplicating discoverable information raise inference costs 20% while dropping task success 2-3%. ETH Zurich (Gloaguen et al.) measured 2.5x tool over-invocation from redundant context. Agents comply with every instruction indiscriminately. Mitko Vasilev Vasilev instrumented Claude Code via local vLLM: 96% of 47M prompt tokens were prefix-cache hits. The compute was cheap. Context management was the real expense. IBM Research's trajectory-informed memory (Fang et al., arXiv:2603.10600) demonstrates what intelligent context management produces. The system converts execution traces into structured tips (strategy, recovery, optimization) and retrieves them via LLM-guided selection. +14.3 percentage points on scenario goal completion for unseen tasks. +28.5pp on the hardest problems. Context selection outperformed context volume. Three production architectures converge on the same principle. Sleep-time compute from Letta and UC Berkeley pre-processes context during idle periods, achieving 5x test-time reduction while maintaining accuracy. @Anthropic's progressive disclosure loads ~100 tokens at Level 1, under 5K on trigger, full resources on demand. Graph-based retrieval structures context as traversable relationships, enabling multi-hop reasoning instead of flat token dumps. The practitioners have already shifted. Boris Cherny 's three-tier context architecture mirrors CPU cache hierarchies: hot tier (150 lines, every session), warm (on-demand via links), cold (search and memory systems). GitHub Copilot validates memory citations at retrieval time. Karthik Ravindran Ravindran orchestrates five modalities (glossaries, ontologies, semantic models, knowledge graphs, digital twins) through a Conductor pattern. The hardware constraint makes this permanent. HBM and DRAM shortages cap how many tokens physically flow to the model per second. Longer windows without intelligent management produce "context rot" regardless of advertised capacity. The solutions fragmenting across the industry must converge on structured context management or each hits the same physical wall. Stop adding context. Start managing it. Graphs provide the structure. Progressive disclosure controls cost. Evaluation determines what survives.
33
7 Kommentare -
Brando Lubis
AMD • 2596 Follower:innen
🚀 Breakthroughs in AI shouldn’t be limited by closed ecosystems. That’s why I’m excited about the AMD ROCm platform — an open and developer-first ecosystem built to remove integration barriers and accelerate AI innovation. With ROCm, developers have the flexibility to build and scale AI solutions without being locked in. Future-ready, adaptable, and open — this is how we advance AI together. #AMD #ROCm #AI #OpenEcosystem #FutureReady #TogetherWeAdvanceAI @AMD #AMDBrandAmbassador
12
2 Kommentare -
Régis Cazenave
S2 Grupo • 10.735 Follower:innen
Former Google CEO Eric Schmidt breaks down the next big leap in #artificial_intelligence and why it will arrive faster than most people think. He explains how modern #AI is racing up the “capability ladder” with new models arriving every twelve to eighteen months, each one more powerful than the last. Schmidt also walks through three forces that will reshape everything from medicine to climate science. Put together, these shifts point toward a world where millions of #agents collaborate, code, and discover at a scale humans alone could never reach. #Innovation #AI #Deeptech https://lnkd.in/eBqB7TMY
17
-
Roberto Hortal
Wall Street English • 5937 Follower:innen
AI can accelerate interview synthesis, but high-quality interviews remain the backbone. Two-step synthesis: master single interviews first, then cross-interview synthesis, with AI as a partner, not a substitute. Rich stories fuel real opportunities. Explore: https://buff.ly/WUBoIrq #ProductManagement
-
Shekhar Bhartiya
Ingentic AI • 10.534 Follower:innen
I coined a new AI paradigm. And then I built it. Four years of doctoral research. Two published papers. Two live platforms. One question that wouldn't leave me alone: What comes after #Agentic #AI? The answer: I propose and coined 'Ingentic AI'. From Latin ingenium — the shared root of both "engine" and "ingenuity." Generative AI generates. Agentic AI acts. Ingentic AI specialises and composes. The core idea: intelligence doesn't require one massive model. It emerges from the structured composition of purpose-built specialist engines — each with its own methodology, its own epistemology — communicating through defined interfaces. I = (E, Φ, Ω) Where E = specialist engines, Φ = composition, Ω = orchestration. 🔬 The Research Two peer-accessible papers now live on Zenodo: 📄 Paper 2 — Methodological foundations of Ingentic AI: Building specialist engines from doctoral research. https://lnkd.in/g7UNxRmr 📄 Paper 1 — Ingentic AI: Intelligence through structured composition of specialist engines. https://lnkd.in/gwn_GjAH Both grounded in doctoral research at Aston Business School — 69 firms, 3,956 quarterly observations, 21 financial KPIs. The Proof 🔵 https://lnkd.in/gWYYGGe2 — Live today. Five specialist engines compose to deliver Big 4-grade business diagnosis. No single LLM — however large — could replicate what these engines do together. 🟣 ingentic.ai — Coming soon. The open paradigm. 💡 Why this matters now Satya Nadella recently said SaaS is dead — that AI agents will replace traditional software. I believe he's directionally right, but the framing is incomplete. #SaaS won't just become agents. SaaS will become engines. Every vertical — CRM, ERP, ITSM, HCM — is really a composition of domain-specific capabilities that should be independently trainable, auditable, and replaceable. Monolithic #SaaS is the mainframe of our generation. This isn't theory. I've spent 25+ years inside enterprise technology — IBM, Accenture, SAP, and now NTT DATA — watching organisations struggle with monolithic systems. Ingentic AI is the architectural pattern I wish we'd had all along. The companies building the foundation models — OpenAI, Anthropic, Google Google DeepMind, Meta , Microsoft, NVIDIA, Mistral AI, xAI — are creating extraordinary raw intelligence. Ingentic AI provides the composition pattern for how enterprises will deploy it: not as one giant model, but as specialist engines that each own their domain. Grateful to my supervisor Dr. Anitha Chinnaswamy for the academic rigour, and to Aston University. Paper 3 is coming: Macro Engines — When SaaS Becomes Ingentic. What do you think — does Ingentic AI name something you're already seeing? I'd love to hear. #IngenticAI #AI #SaaS #EnterpriseAI #Agentic #Research #DBA #Innovation #FutureOfWork #GenAI #LLM #SaaS #ERP Dario Amodei Jensen Huang Arvind Krishna Yann LeCun Sundar Pichai Thomas Kurian Arthur Mensch OpenAI Yutaka Sasaki Abhijit Dubey Christian Klein Andrew Ng
23
4 Kommentare
Weitere Mitglieder, die Thomas Winter heißen
Es gibt auf LinkedIn 760 weitere Personen, die Thomas Winter heißen.
Weitere Mitglieder anzeigen, die Thomas Winter heißen