Job description

Overview Microsoft is reinventing how the world sells. The agents we build power Sales agent (formerly Microsoft 365 Copilot for Sales) for our customers and MSX / Sales Home for Microsoft's own field organization. They help sellers research accounts, prepare for meetings, draft outreach, spot deal risk, and close revenue faster. Our stack spans Azure OpenAI, Microsoft Foundry, Copilot Studio, Microsoft Agent Framework, the Model Context Protocol (MCP), Microsoft Graph, Dataverse, Dynamics 365, Teams, and Outlook. We're looking for a Software Engineer II to join our AI incubation team and work side by side with a Principal engineer on rapid prototyping. You'll take a new agent idea from a whiteboard sketch to a working prototype in days and put it in front of real sellers. If it helps them, you'll help carry it toward production. You'll build agents, MCP servers, tools, and the small UIs that make them demoable, and you'll use AI coding agents every day to move fast. This role suits an engineer who has built real things with LLMs, likes short feedback loops, and is fine throwing away a prototype when the data says it isn't working. Microsoft’s mission is to empower every person and every organization on the planet to achieve more. As employees we come together with a growth mindset, innovate to empower others and collaborate to realize our shared goals. Each day we build on our values of respect, integrity, and accountability to create a culture of inclusion where everyone can thrive at work and beyond. Responsibilities • Turn new agent ideas into working prototypes in days, partnering closely with a Principal engineer, PM, and design from first demo through seller pilot. • Build agents, multi-step workflows, and MCP servers that connect to real seller and customer context in Dynamics 365, Dataverse, Microsoft Graph, and internal sales systems. • Write reusable agent skills, tools, and plugins for Microsoft 365 Copilot and other agent hosts, and contribute them back to our shared platform. • Build lightweight evaluations for your prototypes (golden sets, LLM-as-judge scorers, traces) and report quality, latency, and cost so the team can decide what to scale and what to stop. • Harden the prototypes that prove out with tests, telemetry, authentication, and Responsible AI review, and help hand them off to production surfaces such as Sales Home. • Write short design notes, demo your work to the team and leadership, and share what you learn about new models and agent tooling. • Take part in code and design reviews, and share on-call (DRI) duty for the pilots and services you help ship. Qualifications Required Qualifications • Bachelor's Degree in Computer Science or related technical field AND 2+ years technical engineering experience with coding in languages including, but not limited to, C, C++, C#, Java, JavaScript, or Python OR equivalent experience. Other Requirements • Ability to meet Microsoft, customer and / or government security screening requirements are required for this role. These requirements include, but are not limited to the following specialized security screenings: Microsoft Cloud Background Check: This position will be required to pass the Microsoft Cloud Background Check upon hire / transfer and every two years thereafter. Preferred Qualifications • Bachelor's Degree in Computer Science or related technical field AND 4+ years technical engineering experience with coding in languages including, but not limited to, C#, TypeScript, JavaScript, Python, or Java • OR Master's Degree in Computer Science or related technical field AND 3+ years technical engineering experience with coding in languages including, but not limited to, C#, TypeScript, JavaScript, Python, or Java • OR equivalent experience. • Hands-on experience building applications with large language models (for example prompt design, tool or function calling, retrieval-augmented generation, or agent frameworks), whether in production, internal tools, open source, or substantial personal projects. • Experience building or integrating Model Context Protocol (MCP) servers, or exposing APIs and data to agents as tools. • Experience with one or more agent frameworks or platforms, such as Microsoft Agent Framework, Semantic Kernel, AutoGen, Microsoft Foundry, Copilot Studio, Microsoft 365 Agents SDK, LangGraph, OpenAI Agents SDK, or Claude Agent SDK. • A track record of building fast: hackathon projects, MVPs, demos, or internal tools that went from idea to working software in days, ideally including one that reached real users. • Range across the stack: a back-end API, a simple React/TypeScript front end, and deployment of both to Azure, AWS, or GCP. • Experience evaluating LLM output, such as building small golden datasets, LLM-as-judge scoring, comparing prompts or models, or tracing agent runs (for example with OpenTelemetry). • Working knowledge of Responsible AI practices in real projects: grounding, content safety, prompt injection defenses, and PII handling. • Daily, fluent use of AI coding tools (for example GitHub Copilot, Claude Code, or Cursor) in your development workflow. • Familiarity with Dynamics 365, Dataverse, Microsoft Graph, Power Platform, or CRM systems such as Salesforce, and an interest in sales or customer-facing products. #CEAIjobs Software Engineering IC3 - The typical base pay range for this role across the U.S. is USD $102,100 - $202,200 per year. There is a different range applicable to specific work locations, within the San Francisco Bay area and New York City metropolitan area, and the base pay range for this role in those locations is USD $133,800 - $219,200 per year. Certain roles may be eligible for benefits and other compensation. Find additional benefits and pay information here: careers.microsoft.com This position will be open for a minimum of 5 days, with applications accepted on an ongoing basis until the position is filled. Microsoft is an equal opportunity employer. All qualified applicants will receive consideration for employment without regard to age, ancestry, citizenship, color, family or medical care leave, gender identity or expression, genetic information, immigration status, marital status, medical condition, national origin, physical or mental disability, political affiliation, protected veteran or military status, race, ethnicity, religion, sex (including pregnancy), sexual orientation, or any other characteristic protected by applicable local laws, regulations and ordinances. If you need assistance with religious accommodations and/or a reasonable accommodation due to a disability during the application process, read more about requesting accommodations.

Requirements

Must-haves

This role suits an engineer who has built real things with LLMs, likes short feedback loops, and is fine throwing away a prototype when the data says it isn't working.
Turn new agent ideas into working prototypes in days, partnering closely with a Principal engineer, PM, and design from first demo through seller pilot.
Build agents, multi-step workflows, and MCP servers that connect to real seller and customer context in Dynamics 365, Dataverse, Microsoft Graph, and internal sales systems.
Write reusable agent skills, tools, and plugins for Microsoft 365 Copilot and other agent hosts, and contribute them back to our shared platform.
Build lightweight evaluations for your prototypes (golden sets, LLM-as-judge scorers, traces) and report quality, latency, and cost so the team can decide what to scale and what to stop.
Harden the prototypes that prove out with tests, telemetry, authentication, and Responsible AI review, and help hand them off to production surfaces such as Sales Home.
Write short design notes, demo your work to the team and leadership, and share what you learn about new models and agent tooling.
Take part in code and design reviews, and share on-call (DRI) duty for the pilots and services you help ship.
Other Requirements
This position will be open for a minimum of 5 days, with applications accepted on an ongoing basis until the position is filled.

Nice-to-haves

OR equivalent experience.
Experience building or integrating Model Context Protocol (MCP) servers, or exposing APIs and data to agents as tools.
A track record of building fast: hackathon projects, MVPs, demos, or internal tools that went from idea to working software in days, ideally including one that reached real users.
Range across the stack: a back-end API, a simple React/TypeScript front end, and deployment of both to Azure, AWS, or GCP.
Experience evaluating LLM output, such as building small golden datasets, LLM-as-judge scoring, comparing prompts or models, or tracing agent runs (for example with OpenTelemetry).
Working knowledge of Responsible AI practices in real projects: grounding, content safety, prompt injection defenses, and PII handling.
Daily, fluent use of AI coding tools (for example GitHub Copilot, Claude Code, or Cursor) in your development workflow.
Familiarity with Dynamics 365, Dataverse, Microsoft Graph, Power Platform, or CRM systems such as Salesforce, and an interest in sales or customer-facing products.
Certain roles may be eligible for benefits and other compensation. Find additional benefits and pay information here:

Application prep

Use the role’s language and application details while tailoring your materials.

Keywords to mirror

azureaisketchllmjavajavascriptpythontypescriptapireactawsgcp

Platform: eightfold
Effort: medium