engblogs

Navigating the journey of RAG from proof of concept to production in generative AI space

LLM-as-a-judge framework on Amazon Bedrock Model Evaluation streamlines AI model assessment with automated intelligent evaluation and comprehensive metric categories.

Achieve ~2x speed-up in LLM inference with Medusa-1 on Amazon SageMaker AI

Fine-tune LLMs with synthetic data for context-based Q&A using Amazon Bedrock

An insightful exit interview with JP Phillips shedding light on his journey at Fly.io and the future ahead.

Insights from the IWSLT 2024 Evaluation Campaign by a diverse team of authors

Implementing private federated learning on edge devices for improved accuracy of app selection models while ensuring user data privacy.

Introducing the Databricks AI Security Framework 2.0, facilitating secure and impactful AI deployments with mapped security risks and controls.

Tackling the multifaceted national security threat posed by cybercrime and the collaboration between cybercriminals and state-sponsored groups.

Falcon 3 models now available in Amazon SageMaker JumpStart, offering advancements in open source language models with state-of-the-art performance.

Meta SAM 2.1 vision segmentation model now available in Amazon SageMaker JumpStart for advanced object detection and segmentation workflows

Exploring the intersection of philosophy and AI to navigate computing ethics in a class at MIT