Sign In

Guy Fighel

Co-founder, CTO at SignifAI

Coffee is my passion ☕️, well, Software Engineering too. 👨‍💻 I Love Scaling Infrastructure, Machine Intelligence & Automation.

Highlights

Mar 28 · linkedin

🚀 Just back from KubeCon Europe 2024 and I’m thrilled to share something game-changing: Velocity This platform is a developer’s dream come true. Why? Let’s dive in:

🌟 Quick Start, Impressive Experience: Any developer can get going with velocity.dev in no time, and the experience is super solid!

👨‍💻 Local Development, Staging Power: Imagine coding locally but testing in a production-like staging environment. Deploying all of your services and dependencies on a single Kubernetes cluster.

💡 Real-Time Debugging and On-the-Fly Changes: It’s like local development but in a fully functional, complex environment. Make changes and see immediate results!

🕒 VPs of Engineering, Rejoice: Say goodbye to hundreds of hours lost waiting for the CI pipeline. Velocity.dev streamlines your process significantly.

🔐 For CISOs, A Fortress of Security: Lock down your staging environments and monitor them to prevent any data leakage. You can enforce data and secuirty governance and still let your developers the freedom and velocity.

🤖 Working with this new LLM model? Need hefty GPUs for re-training? With velociy.dev solution, you can use your Jupiter notebook locally and any of your code, while virtually using all of those GPUs on the cluster - complete flexibility in managing and sharing resources.

🎥 Curious to see it in action? Check out this brief video that showcases the power of velocity.dev.

https://lnkd.in/d8RrNiDY

#KubeConEurope2024 #velocitydev #DeveloperTools #InnovationInTech #LLM

Mar 22 · linkedin

An open-source Python library for fast, quality development with LLMs has added support for using Anthropic’s Claude 3 models and its XML function calling capabilities, including nested schemas and def/ref resolution. Example: https://lnkd.in/de8Tyi-S In the example they have created a TaskExtractor class that can extract a fairly complex, nested TaskDetails schema. Haiku does a fantastic job extracting the schema, correctly extracting everything from the task description to due dates to related tasks and metadata. To my knowledge, 🐸 Mirascope is the first open-source library to support this level of nested schema definition and automatic data extraction with Claude 3. Can probably think of other use cases this unlocks, especially with the 200K token context limit.

https://lnkd.in/dxafxhis

Guy Fighel
Add to my network

Experience

Co-founder, CTO at SignifAI

Location

United States