Inspiration
We wanted to remove the friction between understanding requirements and actually shipping them. Teams already track work in YouTrack and build in IntelliJ, so the idea behind YouTrack Boost was simple: bring AI-powered task generation, team-aware planning, and execution support directly into the IDE instead of forcing people to jump between tools.
What it does
YouTrack Boost is an IntelliJ plugin that turns raw requirements into actionable YouTrack tasks. Users can upload a requirements file or record a voice note, and the plugin transcribes and analyzes it with AI to generate structured technical tasks. It also syncs with YouTrack projects, shows a live board inside IntelliJ, suggests assignees based on team skills and workload, allows editing/updating/deleting issues, and even includes an agent-based build assistant that can propose code changes from a selected task.
How we built it
We built the plugin in Kotlin on top of the IntelliJ Platform SDK, using Gradle and JetBrains UI components for the in-IDE experience. We integrated the YouTrack REST API to fetch projects, issues, custom fields, and team members, and used Groq models for requirement analysis plus Whisper transcription for voice input. We also used PDFBox to parse uploaded PDFs and connected an OpenAI-powered agent flow to turn tasks into proposed code changes inside the IDE.
Challenges we ran into
One of the biggest challenges was mapping AI-generated output to real YouTrack project schemas, since fields like priority, state, assignee, and estimation can vary between projects. We also had to keep the UX smooth inside IntelliJ while handling async API calls, file parsing, audio recording, transcription, and issue sync. Another challenge was making task assignment feel useful rather than random, so we added team context, skills, and current workload into the prompt logic.
Accomplishments that we're proud of
We’re proud that YouTrack Boost is not just a concept but a working end-to-end plugin. It can ingest requirements, generate structured tasks, sync them to YouTrack, visualize them in a board, and help distribute work across the team. We’re also proud of the voice input flow and the agent build feature, because they push the plugin beyond project management into actual developer acceleration.
What we learned
We learned that the real value of AI in developer tools is not just generation, but context. The output gets much better when the model knows the project, the allowed YouTrack fields, the team members, and the workload distribution. We also learned a lot about building polished experiences inside IntelliJ, especially around threading, API integration, and designing workflows that feel native instead of bolted on.
What's next for YouTrack Boost
Next, we want to make the agent build flow more powerful and safer, with better previews and one-click apply for code changes. We also want to improve team intelligence with smarter workload balancing, richer skill profiles, and sprint-aware planning. Beyond that, we’d love to support more sources of requirements, tighter YouTrack workflows, and eventually publish YouTrack Boost as a polished plugin others can install and use.
Log in or sign up for Devpost to join the conversation.