Madhu Kumar

[Amadis Technologies @cloudcadi]

𝐂𝐥𝐨𝐮𝐝𝐓𝐞𝐜𝐡 𝐓𝐚𝐥𝐤𝐬 -𝐒𝐞𝐚𝐬𝐨𝐧 𝟐- 𝐄𝐩𝐢𝐬𝐨𝐝𝐞 𝟐- 𝐈𝐬 𝐋𝐢𝐯𝐞!
In this #podcast episode, Alessandro Citrigno, CTO, Nexi Group, shared his experience in implementing successful finops practices for a large PayTech company - Nexi. Hosted by @Sindhu_Priya_D.

[Amadis Technologies @cloudcadi]

Happy to share that Benyl Ross is representing Amadis Technologies in the event this year. We are excited to be part of this as 'Cost' is one of the focus areas this time (that aligns perfectly with our mission!). #lhif #nasscomevents #cloudcosts

[DavidLinthicum @DavidLinthicum]

So many people have asked me to create this, so here it is: 5 minutes of no-nonsense training on what agentic AI is and how to use it correctly.

There’s a lot of noise, confusion, and hype around agentic AI right now. My goal was simple: cut through all of that and provide a clear, practical explanation that gets straight to the point.

If you don’t have a lot of time but need to get smart on this topic quickly, this is where you need to start.

Watch here: https://t.co/1Vibl4WQ8l

#AgenticAI #AI #ArtificialIntelligence #EnterpriseAI #DigitalTransformation #TechLeadership

Agentic AI Explained in Just 5 Minutes
https://t.co/1Vibl4WQ8l

[DavidLinthicum @DavidLinthicum]

AI’s Next Enterprise Cost Trap Is Already Here

This article points to a pattern we’re going to see much more often: companies moving aggressively to replace people with AI and AI agents, only to discover that when they truly price these systems out, the economics often make far less sense than expected.

In many cases, the AI ends up costing more than the people it replaced, while not doing the job as well. That’s the part many organizations still don’t want to confront. The demos look great. The promises are compelling. The board loves the story. But the operational reality is much different when these systems are deployed at scale across the enterprise.

A big part of the problem is overdependence on remote LLMs for everything and anything. Enterprises know they need to leverage AI, and many assume the best approach is to connect as much work as possible to the newest and most powerful external models. What they’re missing is that token consumption is money consumption. Every interaction, every workflow, every agent loop, every automation run burns budget. That spend compounds quickly, and it takes resources away from other priorities the business still needs to fund.

That’s how enterprises fall into a cost trap. They don’t fully understand it yet because they are still focused on capability, not economics. But eventually the bills show up, and leadership begins to realize they created another layer of expensive technical debt.

There is a better option for many use cases: internal AI systems. Models customized for specific enterprise needs, running on infrastructure the company already owns or directly controls, can require a meaningful upfront investment. But once in place, those systems are often far less expensive to operate over time than endlessly paying token-based fees to external providers. They may be a bit slower. They may not always be as flashy as the latest frontier model. But they can get the job done, keep costs predictable, and provide much better long-term control.

What’s surprising is how rarely this option is seriously considered. I understand the appeal of large hosted models. They provide real value. But the reality is that at some point many of these services become too expensive relative to the value they deliver. Enterprises need to stop assuming the newest external model is automatically the right answer for every workload.

I’ve seen this before. The same mistakes many companies made with public cloud providers 10 to 15 years ago are now being repeated with AI. The difference is that this wave of technical debt could end up being 10 to 20 times more expensive.

Enterprises need to wise up. AI absolutely has a place in the business. But if cost, control, and long-term value matter, internal alternatives need to be part of the strategy.

Are companies building real AI advantage right now, or just locking themselves into the next major enterprise cost problem?

Love to hear your thoughts below.

https://t.co/0TggTy01mP

#AI #EnterpriseAI #LLM #AIAgents #GenAI #CloudComputing #FinOps #ITStrategy #CIO #CTO #EnterpriseArchitecture #MachineLearning #DigitalTransformation #TechDebt

[DavidLinthicum @DavidLinthicum]

**Agentic AI Needs to Go on an LLM Diet**

We need to have an honest conversation about how we’re building AI agents.

Too many agentic AI architectures I’m seeing today are completely dependent on calling a large language model every time the agent needs to reason, decide, classify, summarize, route, or respond. Yes, it works. But in many cases, it’s the architectural equivalent of building a rocket ship to bring your groceries home from the market.

The reality is that a large percentage of AI agents don’t need a massive general-purpose LLM sitting behind every action. They need a small, focused, purpose-built model that understands the narrow role the agent is supposed to perform. In many cases, a small language model or highly specialized model can do the job faster, cheaper, and with far less operational risk.

This matters.

When every agent depends on a remote generative AI API call, you introduce latency, cost, unpredictability, governance issues, and a single point of failure. We’ve already seen major AI services go down multiple times over the past year. If your agent stops working every time an external model endpoint has an outage, you may not have built an intelligent system. You may have built a dependency chain with a chatbot attached to it.

The better pattern is the minimum viable AI agent: an agent that does exactly what the business requires, no more and no less. If it needs a large model, use one. But if a smaller, local, domain-specific model can handle the task, that should be the default consideration.

We used to call this good software engineering.

We optimized for the workload. We matched the architecture to the business purpose. We cared about cost, resilience, performance, maintainability, and operational simplicity. Somewhere along the way, with generative AI, we started throwing GPUs, memory, tokens, and giant foundation models at problems that often require something far more precise and far less expensive.

Agentic AI does not get better simply because we attach a bigger model to it. In many cases, it gets slower, more fragile, more expensive, and harder to govern.

The future of AI agents is not just about more powerful models. It’s about better architectural judgment.

Use the LLM when the LLM is required. But don’t make it the default brain for every agent, every task, and every workflow.

Agentic AI needs to go on an LLM diet.

What do you think?

#AgenticAI #ArtificialIntelligence #EnterpriseAI #AIArchitecture #SoftwareEngineering #CloudComputing #GenerativeAI #LLM #AIEngineering #MLOps #AIGovernance #TechnologyLeadership

[Amadis Technologies @cloudcadi]

Here is a quick look to our Monthly Fun Activities!

This month - It is 'Hold it Together Challenge'. Congratulations to the winning team - Shijo S, Kinjal Prisha, Alan Pramil, Sachin S and all the participants.

#funactivities

[Amadis Technologies @cloudcadi]

Think your cloud infrastructure is perfectly optimized? Your Kubernetes nodes might tell a different story!
Discover how CloudCADI can optimize your Kubernetes environment: https://t.co/XT5vC0bfnf

[Amadis Technologies @cloudcadi]

At Amadis Technologies, we reached a point where our ideas outgrew their walls. That’s why we are incredibly excited to announce the official opening of our brand-new '𝐀𝐦𝐚𝐝𝐢𝐬𝐢𝐚𝐧'𝐬 𝐓𝐡𝐢𝐧𝐤 𝐒𝐩𝐚𝐜𝐞'! #officeexpansion #celebration

[Amadis Technologies @cloudcadi]

CloudTech Talks - Season 2 - 1st Episode is out. Hear it all that made T-Mobile, one of the largest telecom enterprises, save a huge chunk of cloud expenses. Listen to a candid conversation with Shalini Aggarwal from T-Mobile hosted by Madhu Kumar. https://t.co/g9W9anLskc

[Amadis Technologies @cloudcadi]

Level Up: 𝐂𝐥𝐨𝐮𝐝𝐓𝐞𝐜𝐡 𝐓𝐚𝐥𝐤𝐬 𝐒𝐞𝐚𝐬𝐨𝐧 𝟐 is Here!

The "Go-To Cloud Podcast" from Amadis Technologies has officially evolved.

Stay tuned till next week for the first episode. #cloudfinopspodcast

[Amadis Technologies @cloudcadi]

Last week, as part of our monthly 'Fun Activities and Celebration', we conducted - "Quick Pick Challenge" for our team to pause, enjoy, rejuvenate and come back refreshed to their desks.

Congratulations winners - Aswin Raj (1st place), Vinotha M P (2nd) and Padmanabhan M (3rd)

[Amadis Technologies @cloudcadi]

That’s a Wrap!
Stay tuned for more updates, and feel free to reach out if you have any follow-up questions from the sessions!

Link for recording of all sessions: https://t.co/afyyx12CZE

[Amadis Technologies @cloudcadi]

This is it! Tomorrow, we wrap up our #GenAI Webinar Series with our most technical and value-packed session yet: "Data to Decisions: Building Intelligent, Scalable Business Systems with AI & Automation."

#webinar Register now: https://t.co/afyyx12CZE

[HH Sheikh Mohammed @HHShkMohd]

Under the directives of the President of the UAE, we launch a new government model. Within two years, 50% of government sectors, services, and operations will run on Agentic AI, making the UAE the first government globally to operate at this scale through autonomous systems.

AI is no longer a tool. It analyses, decides, executes, and improves in real time. It will become our executive partner to enhance services, accelerate decisions, and raise efficiency.

This transformation has a clear timeline. Two years. Performance across government will be measured by speed of adoption, quality of implementation, and mastery of AI in redesigning government work.

We are investing in our people. Every federal employee will be trained to master AI, building one of the world’s strongest capabilities in AI-driven government.

Implementation will be overseen by Sheikh Mansour bin Zayed, with a dedicated taskforce chaired by Mohammad Al Gergawi driving execution.

The world is changing. Technology is accelerating. Our principle remains constant. People come first. Our goal is a government that is faster, more responsive, and more impactful.

[Amadis Technologies @cloudcadi]

1 Week to Go: Master GenAI Token Efficiency!
Register now: https://t.co/iXCl2LZ75W

#webinar #genai

[Amadis Technologies @cloudcadi]

Public #cloudinvestments are expected to soar by 21.5% by 2027- says Gartner!

CloudCADI is built to be the "linchpin" of your strategy, turning complex cloud data into actionable savings.

Get your Free Trial Today - https://t.co/3tVGS3R4DC

[Amadis Technologies @cloudcadi]

New Date, Same Deep Dive: Mark Your Calendars for April 28th!

We are shifting the date for our final session in the #GenAI #Webinar Series from 16th April to ensure we deliver the most comprehensive look at building intelligent, scalable systems.

[Amadis Technologies @cloudcadi]

We believe you can't optimize what you don't fully understand. That's why every Amadis engineer is 100% certified across Azure, AWS, GCP or OCI - no exceptions.

#cloudfinops #cloudcost #cloud

[Amadis Technologies @cloudcadi]

Our Greatest Asset Isn’t Our Tech - It’s Our People!

At Amadis Technologies, we believe that a healthy team is a happy and thriving team. As part of our commitment to employee welfare, we recently organized a comprehensive Health Checkup for our entire workforce.

[Amadis Technologies @cloudcadi]

The definition of FinOps by FinOps Foundation has quietly evolved multiple times - and if you read between the lines, it tells the whole story of where the industry is heading. #finops