Stack Overflow’s Post

Tests are coming to help developers build higher-quality software, but how do we measure the quality of the tests themselves, whether at the PR level or at scale?   Learn more in our latest blog about how Early’s product, using GPT-4o, generated close to 1,000 high quality, working unit tests on ts-morph, a famous OSS project, creating over 17,000 lines of test code, 60% coverage, all in under an hour.   Helping developers code smarter and launch faster is Early’s mission, and every day we are getting one step closer to that goal. This week, we made 17,000 small steps forward.   Find a link to the blog on the first comment. The tests are available on our public GitHub if you want to check them out.

Did you know developers spend over 17 hours a week on maintenance tasks like debugging and refactoring? 😱 AI-generated documentation could be a game-changer in reducing grunt work and improving code quality. We examine some of the challenges of maintaining effective documentation and explore how AI-powered tools could lend a hand in tackling technical debt: https://lnkd.in/eFU5nKfi

In the last months, we've spent a lot of time thinking about how AI can improve the lives of engineering team members. We're working on a tool that turns the English language prompts, into fully adaptable software tests. We're calling it Testdriver 🏎️ Watch the demo below and get in touch if you want to explore more about the functionalities of our AI agent! ✅ Easy to use: Write and maintain tests with natural language alongside your code, inside your versioning system (Github etc.) ✅ Context-aware: Operates independently the operating system (in a VM) with the support of state-of-the-art LLMs and VLMs ✅ Re-usable: Generate and run the same test as many times as you want (with self-healing) ⌛ Here's a short demo I put together:

🎈 New Fine updates🎈 🟣 See when the AI's dev environment is ready 🟣 Built in File Navigator - open files and add them to the context without leaving Fine 🟣 Make manual edits to the code in your browser 🟣 Rollback to earlier stages in the conversation Happy coding 😁

Our recent experiment with GitHub's Copilot AI revealed a significant boost in developer satisfaction, scoring an NPS above 60! Developers enjoyed the ease of coding and tackling bugs efficiently. But that's just the tip of the iceberg. Dive deeper into our findings on how AI tools like Copilot are changing software development: https://lnkd.in/dEB4Ux85

Here are the nine reasons users and stakeholders have told us why they care about how much GenAI is in their codebase, i.e. GenAI Code Transparency. The most common reason: PE-backed companies, and companies who wish to take on later-stage investment or exit, expect to be diligenced on their GenAI code usage. It's just like they are diligenced on Open Source. And they want to stay ahead of any surprises.

Over the last year, we've engaged in extensive discussions with developers, engineering leaders, and executives from organizations worth hundreds of billions of dollars about using GenAI in the SDLC.   A common theme emerged: the importance of GenAI Code Transparency – knowing how much code in a codebase originated from a GenAI tool.   In our latest blog post, we explore nine reasons why developers and organizations care about GenAI Code Transparency.    The top reason? Investor-backed companies expect to have to explain their use of GenAI in future technical due diligences—just like they do today for Open Source usage.   Read the full blog post to learn more about each reason and how GenAI Code Transparency can benefit your organization.   Read more here: https://lnkd.in/dyApKkiD

Are you and your team prepared for the future of software development? Join us on April 18th at 5:00 PM CET for an insightful webinar exploring the intersection of AI code generation and CI/CD automation. CircleCI’s, Olaf Molenveld and Tabnine Brandon Jung, will delve into the nuances of AI-powered coding, discussing both the potential benefits and the important considerations for implementation. Whether you’re already utilizing AI code assistants or considering integrating them into your workflow, this webinar will provide valuable insights to help you maximize efficiency and mitigate risks. https://hubs.li/Q02rJ0V50

Take ur mind off all that election noise!! Are you working on AI - uh - who isn't right :D ? Accelerate & make it easier for your team with this latest release - take 2 secs to check out their tutorial & u may just become the latest hero!