Learn more about our vision and strategy around deploying agents internally at Microsoft.<\/a>}<\/p>\nThe Azure FinOps Budget Agent<\/p>\n
Our Azure FinOps Budget Agent is a perfect example of a scenario for API-based agents.<\/p>\n
The team responsible for managing our Microsoft Azure budget for IT services was looking for ways to reduce costs by 10\u201320 percent. To do that effectively, service and finance managers needed the ability to track their spending quickly, accurately, and easily.<\/p>\n
The conventional approach to solving this problem would be creating a dashboard with access to the relevant data. The problem with a UI-based approach is that it tends to cater to more specific personas by providing data only they need while oversaturating others with information that\u2019s irrelevant to their work.<\/p>\n
\u201cAzure spend is basically the lifeline for our services,\u201d says Faris Mango, principal software engineering manager for infrastructure and engineering services within Microsoft Digital. \u201cGetting the information you need in a concise format that provides a nice, holistic view can be challenging.\u201d <\/p>\n
With the advent of generative AI and Microsoft 365 Copilot, the team knew that a natural language interface would be much more intuitive. The result was the Azure FinOps Budget Agent.<\/p>\n
The team created the agent and the necessary APIs using Microsoft Visual Studio Code. Its tables and functions run on Azure Data Explorer, allowing the APIs and their consumers to access data almost instantaneously, thanks to its low latency and rapid read speeds.<\/p>\n
The tool retrieves data by running Azure Data Factory pipelines that pull and transform data from three sources:<\/p>\n
Our SQL Server for service budget and forecast data<\/p>\n
Azure Spend for the actual spending amounts<\/p>\n
Projected spending, a separate service stored in other Azure Data Explorer tables<\/p>\n
Processing the information relies on our business logic\u2019s join operations, followed by aggregations by fiscal year and service tree levels. These summarize the data per service, team group, service group, and organization.<\/p>\n
After the back end processes the day\u2019s data, it ingests the information into our Azure Data Explorer tables, which the agent accesses by calling via Kusto functions (the query language for Azure Data Explorer). The outcome is very low latency. Typically, the agent returns results in under 500 milliseconds.<\/p>\n
For users, the tool is stunningly simple. They simply access Copilot and navigate to the Azure FinOps Budget Agent.<\/p>\n
The agent provides three core prompts at the very top of the interface: \u201cMy budgets,\u201d \u201cService budget information,\u201d and \u201cService group budget information.\u201d Clicking on one of these pre-loaded prompts returns role-specific information around budget, forecasts, actuals, projections, and variance, all at a single glance. The interface even includes graphs to help people track spending visually.<\/p>\n
If users are looking for more specific information, they can input their own queries. For example:<\/p>\n
\u201cGet me the monthly breakdown of service Azure Optimization Assessment analytics.\u201d<\/p>\n
\u201cFind me the service in this tree with the highest budget.\u201d<\/p>\n
\u201cShow me the Azure budget for our facilities reporting portal.\u201d<\/p>\n
\u201cWhich service deviates most from its budget forecasts?\u201d<\/p>\n
The Azure FinOps Budget Agent primarily serves two groups: service managers who directly oversee spend for Azure-based services and FinOps managers responsible for larger budget silos.<\/p>\n
Mango is responsible for the internal UI that helps network employees access parts of the Microsoft network. With 18\u201320K users per month, budgeting and forecasting are highly dynamic due to traffic fluctuations and the resourcing that supports them. He also oversees the internal portal that helps service engineers manage our networks. The tool is growing rapidly as we onboard more teams, so forecasting is anything but linear.<\/p>\n
For both of these services, keeping close track of spending is essential. Mango finds himself checking the Azure FinOps Budget Agent about twice a month to gauge how his services are trending.<\/p>\n
\u201cIt\u2019s taking me less time to do analysis and come up with accurate numbers. And the enhanced user experience just feels more natural, like you\u2019re asking questions conversationally rather than engaging with a dashboard.\u201d<\/p>\n
![\"A](\"https:\/\/www.europesays.com\/ai\/wp-content\/uploads\/2026\/04\/Faris-Mango.png\")
<\/p>\n
Faris Mango, principal software engineering manager for infrastructure and engineering services, Microsoft Digital<\/p>\n
For FinOps managers, the value is more high-level. They are responsible for overseeing tens of services featuring vast volumes of Azure usage across storage and compute while managing strict budgets. That requires constant vigilance.<\/p>\n
Switching context from one dashboard to another to track different Azure management groups was a constant hassle for them. Now, they use the Azure FinOps Budget Agent to get an up-to-date view of the overall spend picture. It gives them a place to start. From there, they can drill down if he sees any abnormalities.<\/p>\n
\u201cIt\u2019s taking me less time to do analysis and come up with accurate numbers,\u201d Mango says. \u201cAnd the enhanced user experience just feels more natural, like you\u2019re asking questions conversationally rather than engaging with a dashboard.\u201d<\/p>\n
The arrival of the Azure FinOps Budget Agent is just one example of how agents take your context and get your people the answers they care about faster at less cost.<\/p>\n
Benefits like these are spreading across teams throughout Microsoft. Overall, we\u2019ve been able to save 10\u201312 percent of our overall Azure cost footprint for Microsoft Digital, and individual users are thrilled at the amount of time and effort they\u2019re saving.<\/p>\n
\u201cNow the info is at people\u2019s fingertips. The advantage of an agent is that users don\u2019t have to understand a complex UI, so they can get quick answers and get back to work.\u201d<\/p>\n
![\"A](\"https:\/\/www.europesays.com\/ai\/wp-content\/uploads\/2026\/04\/Shedab-Beg.png\")
<\/p>\n
Shadab Beg, principal software engineering manager, International Sovereign Cloud Expansion<\/p>\n
Five key strategies for building an API-based agent<\/p>\n
After seeing what we\u2019ve accomplished with API-based agents, you might be wondering how to put them into action at your organization. This step-by-step guide can help you get there.<\/p>\n
Building an API-based agent needs to fulfill multiple requirements. It has to expose APIs, align with real user needs, integrate seamlessly with Microsoft 365 Copilot, and work reliably, efficiently, and scalably. Achieving those outcomes depends on five key strategies. <\/p>\n
![\"\"](\"https:\/\/www.europesays.com\/ai\/wp-content\/uploads\/2026\/04\/One-icon.png\")
<\/p>\n
\nStart with user intent, not the API\n<\/p>\n
Start by asking a simple but powerful question: What will users actually ask your agent? Instead of designing the API first, flip the process:<\/p>\n
Gather real user queries to understand actual use cases.<\/p>\n
Refine the queries using prompt engineering techniques to align them with expected AI behavior.<\/p>\n
Design the API to provide structured responses to those refined queries.<\/p>\n
By starting with user intent, you ensure your agent answers real user questions directly, avoids over-engineering unnecessary endpoints, and delivers meaningful results without excessive back-end processing.<\/p>\n
\u201cNow the info is at people\u2019s fingertips,\u201d Beg says. \u201cThe advantage of an agent is that users don\u2019t have to understand a complex UI, so they can get quick answers and get back to work.\u201d<\/p>\n
The advantage of an agent is that users don\u2019t have to understand a complex UI, so they can get quick answers and get back to work.\u201d<\/p>\n
Key learning: An API that doesn\u2019t align with user intent won\u2019t be effective\u2014even if you design it well.<\/p>\n
![\"\"](\"https:\/\/www.europesays.com\/ai\/wp-content\/uploads\/2026\/04\/Two-icon.png\")
<\/p>\n
\nDesign APIs for Microsoft 365 Copilot Integration\n<\/p>\n
It\u2019s important to build an API schema that returns precise and structured data to make it easy for Copilot to consume. This ensures your APIs return data in a format that directly answers user queries. Copilot expects responses in under three seconds, so focus on optimizing API responses for low latency.<\/p>\n
Once you have your list of key questions, design your API schema to return the exact data you need to answer those questions. Your goal should be to ensure every API response has a structure that makes it easy for Copilot to understand.<\/p>\n
![\"\"](\"https:\/\/www.europesays.com\/ai\/wp-content\/uploads\/2026\/04\/Three-icon.png\")
<\/p>\n
\nTeach Microsoft 365 Copilot to call your API\n<\/p>\n
Copilot needs to know how to call your API. Manifests and OpenAPI descriptions accomplish that training. <\/p>\n
Create detailed OpenAPI documentation and plugin manifests so Copilot knows what your API does, how to invoke it, and what responses to expect. You\u2019ll likely need to adjust to these files through a process of trial and error. \u00a0<\/p>\n
![\"\"](\"https:\/\/www.europesays.com\/ai\/wp-content\/uploads\/2026\/04\/Four-icon.png\")
<\/p>\n
\nScale APIs for performance and reliability\n<\/p>\n
Once you have your schema and integration in place, it\u2019s time to move on to the primary engineering challenge: making your API scalable, efficient, and reliable.<\/p>\n
Prioritize the following goals:<\/p>\n
Fast response times: Copilot expects quick answers.<\/p>\n
High scalability: This ensures seamless performance at scale.<\/p>\n
Reliable uptime: The system needs to remain robust.<\/p>\n
We recommend setting a very strict latency limit while implementing your API to retrieve data, since Copilot needs time to generate its response. Existing API endpoints often involve complex data joins rather than simply returning rows from data tables. This complexity can lead to longer processing times, particularly with intricate queries that involve multiple data stores.<\/p>\n
To address these potential delays, pre-cache results to significantly enhance performance. This can help overcome the latency requirements imposed by Copilot.<\/p>\n
At this point, you\u2019ll see why starting with user intent and iteratively refining API design is important. By grounding your work in user behaviors, you\u2019ll align with the following best practices:<\/p>\n
Structure your response to directly address user queries.
Instead of just returning raw data, the API should provide meaningful insights Copilot can interpret. Prompt engineering marries user intent with the most understandable API schema.<\/p>\n
Keep your API flexible enough to adapt to evolving business needs.
Real-world workflows change over time, and an API should be able to support those changes without massive refactoring.<\/p>\n
Avoid performance bottlenecks caused by unnecessary complexity.
Understanding the exact data requirements up front prevents heavy joins, excessive filtering, and inefficient data retrieval logic.<\/p>\n
Optimize for Copilot\u2019s real-time response constraints.
With a strict limit on latency, consider pre-optimization techniques like pre-caching results and simplifying query logic from the very beginning of your API implementation.<\/p>\n
If you attempt to build a scalable, reliable API without first understanding how users will interact with your agent, you\u2019ll spend months reworking the schema, debugging inefficiencies, and struggling with integration challenges.<\/p>\n
Key learning: A fast, scalable, and reliable API isn\u2019t just about technical optimization. It starts with a deep understanding of the questions it needs to answer and how to structure responses so Copilot can interpret them correctly.<\/p>\n
![\"\"](\"https:\/\/www.europesays.com\/ai\/wp-content\/uploads\/2026\/04\/Five-icon.png\")
<\/p>\n
\nConsider compliance and responsible AI\n<\/p>\n
Unlike custom agents or OpenAI API integrations, knowledge-only agents require far less effort to meet Microsoft\u2019s Responsible AI Standard. Microsoft tools\u2019 built-in compliance capabilities handle much of the complexity. As a result, you can focus on efficiency and optimization rather than regulatory hurdles.<\/p>\n
\u201cAgent-based automation must balance speed with responsibility,\u201d Nasir says. \u201cWe embed compliance, cost control, and telemetry from the start, so our systems don\u2019t just scale, they mature.\u201d<\/p>\n
Key learning: It\u2019s helpful to revisit your existing compliance, governance, and responsible AI processes and policies before implementing AI solutions. Copilot adheres to protective structures within your Microsoft technology ecosystem, so this process will ensure you\u2019re starting from the most secure position.<\/p>\n
APIs and the agentic future<\/p>\n
Building API-based agents is more than just an integration exercise. It\u2019s about creating scalable, intelligent, and compliant AI-driven workflows. By aligning your API design with user intent, you set Microsoft 365 Copilot free to retrieve and interpret information accurately. That leads to a seamless AI experience for your employees.<\/p>\n
Thanks to Copilot\u2019s built-in security and compliance features, API-based Copilot agents are some of the most efficient, compliant, and enterprise-ready ways to deploy AI solutions. They represent another step into an AI-first future tailored to your employees\u2019 and organization\u2019s needs.<\/p>\n
Tools like API-based agents democratize the information we all need to do our jobs better, because we\u2019re all getting the same data from the same place. This is why an AI-first mindset is actually human-first.<\/p>\n
![\"\"](\"https:\/\/www.europesays.com\/ai\/wp-content\/uploads\/2026\/04\/Key-takeaways-badge.png\")
<\/p>\n
Key takeaways<\/p>\n
Here are some things to keep in mind when designing agent-powered experiences that are fast, reliable, and aligned with user expectations.<\/p>\n
Response time is key. Choose single APIs that have low latency to facilitate both the technical requirements of Copilot and users\u2019 needs.<\/p>\n
Consider the source. Data has to be high-quality on the backend. It\u2019s worth reviewing your data and ensuring the hygiene is good.<\/p>\n
Agents and APIs need to align. Design with task-centric, well-structured agents. Determine your high-level goals, then use the OpenAI standard, OpenAPI, or graph schemas to describe task endpoints. Define each API\u2019s capability, input schema, and expected outcome very clearly. <\/p>\n
Plan ahead to avoid surprises. Design your APIs to minimize potential side effects, especially through enabling natural-language-to-API mapping, because that\u2019s the biggest change in methodology.<\/p>\n
Design for visibility. Agents need to be observable and explainable, so implement metrics-driven monitoring. Having API-level telemetry in addition to Copilot-level telemetry enables continuous improvement.<\/p>\n
![\"\"](\"https:\/\/www.europesays.com\/ai\/wp-content\/uploads\/2026\/04\/Try-it-out-badge.png\")
<\/p>\n
Try it out<\/p>\n
Create and deploy an agent in Microsoft Copilot Studio.<\/a><\/p>\n![\"\"](\"https:\/\/www.europesays.com\/ai\/wp-content\/uploads\/2026\/04\/Related-links-badge.png\")
<\/p>\n
Related links<\/p>\n
![\"\"](\"https:\/\/www.europesays.com\/ai\/wp-content\/uploads\/2026\/04\/Wed-like-to-hear-from-you-badge.png\")
<\/p>\n
We\u2019d like to hear from you!<\/p>\n","protected":false},"excerpt":{"rendered":"Agentic AI is the frontier of the AI landscape. These tools show enormous promise, but harnessing their power…\n","protected":false},"author":2,"featured_media":13408,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[11],"tags":[24,420,7853,10371,416,10372,9938,320,8172,4528,7852,10373],"class_list":{"0":"post-13407","1":"post","2":"type-post","3":"status-publish","4":"format-standard","5":"has-post-thumbnail","7":"category-microsoft","8":"tag-ai","9":"tag-azure","10":"tag-azure-copilot","11":"tag-azure-networking","12":"tag-copilot","13":"tag-data-and-ai","14":"tag-enterprise-data","15":"tag-microsoft","16":"tag-microsoft-365-copilot","17":"tag-microsoft-azure","18":"tag-microsoft-copilot","19":"tag-network-and-infrastructure"},"_links":{"self":[{"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/posts\/13407","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/comments?post=13407"}],"version-history":[{"count":0,"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/posts\/13407\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/media\/13408"}],"wp:attachment":[{"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/media?parent=13407"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/categories?post=13407"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/tags?post=13407"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}
![\"\"](\"https:\/\/www.europesays.com\/ai\/wp-content\/uploads\/2026\/04\/Engage-with-our-experts_blogs.png\")
<\/p>\n![\"A](\"https:\/\/www.europesays.com\/ai\/wp-content\/uploads\/2026\/04\/Faisal-Nasir.png\")
<\/p>\n