Tag: Azure AI Foundry

Global AI Bootcamp 2025 (Sydney)

Sydney’s AI community came together for an inspiring event at Microsoft Reactor Sydney, drawing over 120 passionate participants. The day was a dynamic mix of thought-provoking sessions and vibrant discussions.

We had a wide range of attendees, including data engineers, architects, business leaders, analysts, student university professors, etc. It was nice to see many familiar faces.

Model Agnostic AI Implementation

The topic of my presentation was “Model Agnostic AI Implementation.“
Synopsis: Choosing a particular AI Model or API can be time-consuming and challenging when designing and implementing an AI solution. Even Azure AI Foundry has over 1800+ AI models, which can be overwhelming. At a point in time, the model’s performance and benchmarking can also change significantly. What might look like the best option now can be superseded by another model in six months with better accuracy and lower transaction costs.
By implementing and extending the Azure AI Model Inference API, we can eliminate the need to select the perfect model upfront and avoid analysis paralysis. In the session, I will dive deep into this API abstraction to speed up the Model selection and implementation process. Later, if we want to change the model for specific scenarios, we will be able to do so without impacting the rest of the applications or systems.

If you are interested in the slides, here is the Link – slides from the presentation

Common Questions

AI Adoption and Strategy:

Several attendees were keen to understand the AI strategy and the steps. Below is the summarised version of what we have discussed. However, Microsoft Cloud Adoption Framerok has a detailed checklist and steps an organisation can follow: https://learn.microsoft.com/en-us/azure/cloud-adoption-framework/scenarios/ai/

Benchmarking

During the session and 1:1 chat afterwards, many people had questions about benchmarking models, including how to independently verify models and where to find benchmarking information. As shown during the presentation, Azure AI Foundry has benchmarking of many popular models, and we can compare the quality/cost based on the tasks/skills.
You can create an account (free) and log in to explore the ranking: https://ai.azure.com/explore/benchmarks
You can learn more about benchmarking here: https://learn.microsoft.com/en-us/azure/ai-foundry/concepts/model-benchmarks

There are countless other sites where you will find benchmarking information about different models. It will boil down to the specific use case you are trying to solve. Some similar sites are:

Learning and Career Path

Several students and technical folks were interested in how they could learn, sharpen their AI skills to be more industry-aligned and help them with AI-related jobs or projects. Some helpful links are below:

Community Posts

If you want to get in touch with some of the other speakers or organisers, visit these LinkedIn posts

The Era of Multi AI Agent

Most organisations have been building bots and agents for the last few years. Some of the agents are now integrated with GenAI to perform complex tasks and serve customers with a more human-centred experience.

What is becoming a clear Multi-AI Agent solution with humans in the loop will have more trust in the outcome and the most adoption by the customer. A multi-AI agent solution will also allow us to break a complex process/task into sub-processes/tasks where each agent will perform a very domain-specific task. There is an overhead of having an orchestration agent that will need to delegate agent-specific tasks. However, breaking down the process will make the agent simpler to implement, more accurate, testable and manageable in the long run.

Before proceeding, I must mention some of the graphics I have reused from Mads Bolaris‘s presentation.

Azure AI Foundry has been a great way to create single or multi-AI agents. Some amazing samples, such as the Azure AI Assistance Tool and Questionnaire Multi-Agent, can speed up the AI assistance implementation.

Another massive benefit of Azure AI Foundry is that it provides the full enterprise package, including data storage, building and testing agents, choice of models, prompt samples/accelerators, and the ability to add custom actions through Azure Function, Open API, etc.

Whether your solution is a single AI agent or a multi-AI agent will depend on the scenario or problem you are solving. If you are building a chatbot or an agent for a very specific task, a single AI agent might be the way to go. However, for most enterprise customers, multi-AI agents will be the solution, and we should apply the Process Framework to break the process into small, manageable chunks.

While a single (complex) AI agent may solve many of the business use cases, we should keep in mind that such an agent will be:

  • More challenging to test and debug issues for different scenarios
  • More likely to hallucinate
  • Harder to measure the quality of the response
  • Complex to build and maintain without impacting sub-functionalities/tasks

To automate and implement business processes, a multi-agent system can be implemented through Orchestrator and Worker agents.

Each worker agent can validate the progress against the goal, check for the quality of output, implement hard stops to iterations, and provide a response when fulfilling the criteria.

For the Muti-agent system, orchestration options are available: AutoGen or Semantic Kernel.

There are some great blogs: Microsoft Semantic Kernel and AutoGen: Open Source Frameworks for AI Solutions, and Microsoft’s Agentic Frameworks: AutoGen and Semantic Kernel covers the option in more detail. However, in a nutshell, if you plan to deploy your multi-agent solution in production in the next few months, Semantic Kernel is the way to go. Semantic Kernel has reached version 1.0 across .NET, Python and Java and is ready for production use. Below are some useful resources and links for both frameworks;

Semantic Kernel

AutoGen