Walk Through Your Personal LLM Tool Stack and Defend Each Choice

Tell me about the AI tools in your own workflow: which large language models or agents do you rely on, and why those over the alternatives? ```hint Organize by job, not by brand Tie each tool to a task you repeat every week, then explain what made it beat the other tools you tried for that task and where it still lets you down. ``` ### Clarifying Questions - Should I cover tools I use at work, in my personal life, or both, and is it fine to name products from companies other than Microsoft? - Do you want only chat assistants, or also agents that take actions for me, such as coding agents or tools that browse and complete multi-step tasks? - How deep should the "why" go: my experience as a user, or the model-level trade-offs behind it, such as reasoning quality, context length, latency, cost and data handling? ### What a Strong Answer Covers - Concrete, recurring jobs that each tool is used for, rather than a list of product names. - The criteria behind each choice (quality on that job, access to the right context, where the tool lives, speed, data protections) and an honest comparison with at least one alternative. - Awareness of limitations such as confident errors, stale knowledge and unreliable multi-step agents, and the habits used to verify output. - Product insight: what the candidate would change about a tool they use, and how that connects to Microsoft's own AI products such as Copilot. ### Follow-up Questions - Take the tool you use most. What single change would make you use it twice as often, and how would you measure whether the change worked? - Describe a time an AI tool's output misled you. What did you change in how you use it, and how should the product have behaved? - An enterprise customer bans all AI assistants over data concerns. How would you go about earning that customer's trust?