NVIDIA Jetson Adds Agentic AI Support for Physical Applications
NVIDIA announced JetPack 7.2 and NemoClaw support at COMPUTEX, expanding agentic AI capabilities for physical-world applications.
ARTICLES BY THEO ALMEIDA
LATEST NEWS
-
-
NVIDIA Introduces Jetson Agentic AI for Physical World Applications
NVIDIA announced JetPack 7.2 and NemoClaw support for Jetson at COMPUTEX, enabling agentic AI deployment in robotics and industrial automation.
-
AWS Introduces Secure Auth Code Flow for AgentCore Gateway with MCP Clients
AWS outlines a secure authentication method for AI assistants using AgentCore Gateway and MCP clients, enabling identity-verified access to enterprise tools.
-
Amazon Bedrock AgentCore Identity Now Supports AWS Secrets Manager Secrets
Amazon Bedrock AgentCore Identity now allows users to reference existing AWS Secrets Manager secrets, providing full control over encryption and access policies. This update supports cross-account secret sharing within the same AWS region.
-
GitHub Copilot shifts to usage-based pricing model
GitHub Copilot users face significant cost increases as the service moves to a usage-based pricing model, with some burning through monthly credit limits in under a day.
-
AMD Introduces Fortran OpenMP Profiling Guide for GPU Optimization
AMD released a new profiling guide for Fortran OpenMP offload applications on AMD GPUs, featuring examples on the MI300A APU with ROCm 7.2 tools.
-
AMD Enables ROLL Framework on GPUs for Reinforcement Learning
AMD has made ROLL, an open-source RL framework, run out-of-the-box on its Instinct GPUs with ROCm software, without code changes.
-
Amazon Introduces Bedrock AgentCore Payments for Secure Agentic Transactions
Amazon Bedrock AgentCore payments, available in preview, enables agents to transact on behalf of users with built-in guardrails to mitigate risks. The service is available in four regions.
-
Amazon Introduces AgentOps to Operationalize Agentic AI with Bedrock AgentCore
Amazon announced AgentOps, a framework to manage agentic AI at scale using Amazon Bedrock AgentCore, with four pillars for governance, operations, evaluation, and observability.
-
Amazon Quick Streamlines Rare Cancer Research with Unified Data Integration
Amazon Quick Research enables faster analysis of rare cancer data by integrating biomedical databases, reducing setup time from weeks to hours. The tool uses AI to generate cited reports with versioned revisions.
-
AWS Expands MCP Support for Amazon Bedrock AgentCore Gateway
AWS announced new capabilities for Amazon Bedrock AgentCore Gateway to enhance enterprise MCP deployments. The update supports dynamic listing, streaming, and OAuth 2.0 token exchange for secure, centralized access.
-
Amazon Quick Integrates MCP for Time-Series Market Insights
Amazon Quick now connects to time-series databases via MCP, enabling financial analysts to query market data using natural language. This integration is demonstrated with KDB-X and allows users to access real-time trade data.