Isaac Levin
Developer Advocate at JetBrains | Host of Coffee and Open Source | Microsoft MVP
Woodinville, Washington, United States
Actions
Isaac Levin is a Developer Advocate at JetBrains and a Microsoft MVP, with over 15 years of experience as a developer for the web, mostly in .NET. Lately he spends a lot of his time building agent workflows for his own projects and figuring out where a human still needs to be the one pressing the button. He hosts Coffee and Open Source, a podcast and live stream where he interviews folks from across the tech industry, with more than 180 episodes since 2020. He has also built apps like PresenceLight, SocialPoster and plenty of things wasting space on GitHub.
Area of Expertise
Topics
180 Conversations: What Six Years of Interviewing Developers Taught Me
Since 2020 I've hosted Coffee and Open Source, a podcast and live stream where I talk with folks from across the tech industry. That's more than 180 episodes so far, which is a lot of yapping.
Now I'm going back through the whole archive and treating it like data. I'm tagging every episode for the themes that come up, like how people got started in open source, how they found contributors, burnout and career changes. Then I'm counting, and looking at what changed between 2020 and now. How often did AI come up in 2020, compared to this year? I'll have the numbers.
I'm especially interested in the setbacks, and how people came back from them. You'll hear those from the guests themselves, in short clips and quotes used with their permission.
There's no code in this one, just what six years of developer stories add up to, in their own words and with real numbers behind it.
Short Links, Big Impact: Automating Social Distribution with .NET and GenAI
We’ve all built a link shortener to learn a new language, but in the age of GenAI and Cloud-native orchestration, a simple redirect isn't enough. What if your link shortener didn't just shorten a URL, but understood its content, crafted a tailored social media campaign, and scheduled the posts automatically?
In this session, we take the classic "Link Shortener" project and supercharge it using the modern .NET stack. We will walk through the journey of building a Social Distribution Engine, a project that serves as the perfect playground for mastering Aspire, Azure OpenAI, and background service orchestration.
We’ll dive deep into the technical architecture, including:
The Intelligent Pipeline: Using Azure OpenAI to scrape destination content, summarize key takeaways, and generate platform-specific social media copy (X, LinkedIn, Mastodon).
Cloud-Native Orchestration: How Aspire simplifies the "wiring" of our web frontend, Redis cache, and background worker services without a single manual connection string.
Scheduled Execution: Implementing reliable background processing to handle rate-limiting and auto-scheduling for social API broadcasts.
Observability by Default: Leveraging the Aspire dashboard to trace an incoming link through the AI generation process to a successful social post.
This isn't just a "how-to" on link shortening; it’s a masterclass in using .NET to build integrated, AI-driven workflows that solve real-world productivity gaps.
I Let an Agent Migrate a .NET Framework App. Here's the Bill.
I've been giving talks about getting apps off .NET Framework for a long time. Lately the approach I've been teaching is to go piece by piece, with YARP in front of the old app, so you can move one part at a time and keep shipping.
A lot has changed. The promise now is that an agent can do a lot of the migration for you. So I'm letting one do it and writing down what it costs.
I'm starting with a sample, that covers Web Forms, MVC and a WinForms app that talks to WCF. Before the agent touches anything, I'm writing characterization tests so we know what working looks like. Then the agent migrates while I supervise and log the hours, the tokens, every time I have to step in, and everything that breaks along the way. I'm keeping a close eye on the usual suspects, like WCF, System.Web, config transforms and all that static state.
Together we will put the agent's version next to the piece-by-piece YARP approach and compare cost and risk. What I really want to know is how far the agent gets on its own, and what the last 20 percent costs. You'll leave with a realistic idea of what to hand an agent on your own migration, and what to keep for yourself.
Build a .NET Terminal UI Live: From Throwaway Console App to Tool You Can Maintain
We've all written the console script that started as ten lines and grew into a tangle of `Console.ReadLine` calls nobody wants to touch. In this one-hour live build, we'll take exactly that kind of app and turn it into a terminal UI that's pleasant to use and easy to maintain.
We'll code together, step by step, and cover:
- Prompts, selections, and validation that guide users through a multi-step workflow
- Tables, panels, and progress bars that show what's happening and what happened
- Separating UI from logic, so the workflow can be tested without scripting a terminal
- Optional bonus: a quick look at wiring in Aspire for diagnostics, if time allows
**Prerequisites:** basic C# and a curiosity about command-line tools. No TUI experience needed.
Expect about 45 minutes of building and 15 for questions, plus a reusable project structure to take home.
Does Your Coding Agent Need an IDE? Semantic Code Tools vs. grep in .NET
A lot of coding agents find their way around a codebase with text search and a shell. Meanwhile the IDE sitting right next to them already knows every caller of every method, and it can rename a symbol across a whole solution without touching the wrong thing. Some tools can now hand that to an agent over MCP, so I wanted to know if it actually helps, and by how much. To be honest, I don't know yet.
I'm running the same .NET tasks three ways. One agent only gets file search and the shell. Another gets IDE tools over MCP. The third gets an open-source setup built on a language server, so anyone can reproduce it without a license. The tasks are the kind text search struggles with, like a rename across a solution, a signature change with a lot of callers, and a bug that only shows up in one test. For every run I'm tracking tokens, time, whether it worked, and how often the build broke along the way.
By the end you'll know which kinds of tasks are worth wiring up semantic tools for, and how to set them up for your own agent. You'll also get the harness, so you can run the same comparison on your own code.
Please note that Sessionize is not responsible for the accuracy or validity of the data provided by speakers. If you suspect this profile to be fake or spam, please let us know.
Jump to top