PyCharm

AI agent costs don’t always explode overnight. Sometimes they creep up quietly: • Extra reasoning steps • Repeated retrievals • Unnecessary tool calls • Verbose prompts • Slow model responses • Workflows that take longer than expected In a demo, this may be invisible. In production, it becomes a budget and user experience problem. That’s why cost and latency should be part of agent evaluation from the start – not something teams discover through surprise bills or user complaints. 👉 Read the full guide to learn which production metrics you should monitor: 🔗 https://jb.gg/qy3a3v 🔗

Attending PyCon Italy this week? Come say hi 👋 We're sponsoring the event in Bologna on May 28–30: 💬 Find our team onsite to talk about the IDE, Python, or just to chat. 🧠 Take our quiz to test your Python knowledge – look for screens around the venue. 🎨 Grab PyCharm stickers before they're gone. ☕ Enjoy a coffee break (our treat!). See you there! #PyConIT #PyConIT26

“What should we actually measure?” is where AI agent evaluation often gets messy. Some metrics check whether the output is accurate and safe. Some check whether a RAG system retrieved the right sources. Some check whether the agent completed the workflow correctly. And some only become visible in production – like cost, latency, and regressions. That’s why one score is not enough. This carousel breaks down 8 metrics and evaluation areas teams can use to make AI agents easier to test, monitor, and improve. Read the full blog post for more details: https://jb.gg/qy3a3v

Turn your webcam into a real-time object detection app with #TensorFlow and PyCharm. In this tutorial by Iulia Feroli, you’ll use a laptop webcam and PyCharm notebook to: 1️⃣ Capture webcam frames. 2️⃣ Convert them into TensorFlow tensors. 3️⃣ Run SSD MobileNet V2 inference from TensorFlow Hub. 4️⃣ Filter detections by confidence score. 5️⃣ Draw bounding boxes with OpenCV. 6️⃣ Validate everything locally. Then you can deploy the same pipeline to #ReachyMini – allowing the robot to follow detected objects, react with its antennas, and show the results on a live dashboard. Code snippets and the full GitHub repo are provided, so you can follow along or adapt your own project. ➡️ Read the tutorial: https://jb.gg/lrnm5i

Before you ship an AI agent, don’t stop at “the answers look good”. Agents can fail in less obvious ways: • Completing one step but missing the actual goal • Choosing the wrong tool • Passing bad parameters • Producing answers that sound right but aren’t grounded • Getting slower or more expensive after updates That’s why production-ready agents need more than standard LLM evaluation. You need to test whether the agent can work – and observe whether it is working in production. The carousel breaks this down into 7 practical checks you can use before putting an AI agent in front of users. 👉 Read the full blog post by Naa Ashiorkor Nortey for more details: https://jb.gg/qy3a3v

What do you enjoy most about coding? For Johannes Rüschel, it’s seeing a problem being solved. He also enjoys that software is never really finished. There’s always an opportunity to improve the design, handle new edge cases, or rethink a solution. Watch the full interview about building better developer tools: 🔗 https://lnkd.in/eShmKdTg 🔗

LLMs don’t just need text. They need context. Parsing PDFs isn’t just extracting words. It’s capturing structure, sidebars, and relationships between sections. That’s where many AI systems still fall short. Watch our Python conference recap: https://lnkd.in/eGxkpafb

Python monorepos can quickly turn into absolute dependency chaos. Here are five ways PyCharm makes working with uv workspaces easier: 1️⃣ Detects when you open a workspace 2️⃣ Sets up or detects the workspace virtual environment 3️⃣ Syncs dependencies directly from pyproject.toml 4️⃣ Supports navigation and refactoring across workspace members 5️⃣ Runs and debugs projects using the workspace interpreter It works with #uv, #Hatch, and #Poetry workspaces, too! 👉 Watch the full video by Paul Everitt to see the full workflow in action: https://lnkd.in/eTdzZcih

“Coding is only a means to an end.” Johannes Rüschel – software engineer, technical coach, open-source contributor, and longtime PyCharm power user – shared his perspectives on software engineering with us. ▶️ Watch the full conversation: 🔗 https://lnkd.in/eShmKdTg 🔗

You can now enable Pyrefly as an external type provider in PyCharm, dramatically increasing the speed of the IDE’s code insight features. Engineered in Rust by Meta, this next generation type checker is built for high performance analysis of large scale Python codebases. By delegating analysis to this faster engine, you receive immediate feedback even as your project grows in complexity. KEY BENEFITS 🔹 ENHANCED CODE INTELLIGENCE: Pyrefly powers essential features like type inference, type-related diagnostics, quick documentation, and inlay hints. 🔹 SCALABILITY: Specifically designed to handle massive Python codebases with high precision and minimal overhead. 🔹 SUPERIOR EFFICIENCY: Its Rust-based architecture delivers significantly faster performance than the built-in engine. 🔹 SEAMLESS SETUP: If you do not have Pyrefly installed yet, PyCharm will handle the installation for you automatically. HOW TO ENABLE IT To start using the Pyrefly engine, go to the Type widget at the bottom of your window. By default, the IDE uses the built-in type engine – simply click the widget and select Pyrefly to switch. Once enabled, you will see a Pyrefly icon in the status bar which you can hover over to verify the active version. Note: The integration currently supports local interpreter configurations, with support for Docker, WSL, and SSH planned for future releases. 👉 Read the blog post for more details: https://lnkd.in/dniVzCYB #PyCharm #Python #Meta #Rust #Programming #LSP #OpenSource