Home›Agents

Agents

Agentic AI systems use LLMs as reasoning engines that plan, use tools, and execute multi-step tasks autonomously. Covers design patterns, orchestration frameworks, and real-world deployments.

322 articles

Agents deep dives from the blog

All posts →
AI Engineering

Multi-Agent Orchestration Patterns: Supervisor, Swarm, Handoff, and When One Agent Is Enough

12 min read · Read →
AI Agents

How Agent Skills Are Reshaping Agentic Workflows in 2026

12 min read · Read →
AI Engineering

How to Build AI Agents That Don't Fall Apart in Production

14 min read · Read →
60 articles
Introducing Gemini 4 Neon
Google DeepMind· 9 min read· Today
Introducing Gemini 4 Neon

Gemini 4 Neon is a new Google DeepMind model that delivers frontier‑level intelligence for complex software engineering, knowledge work, and scientific workflows, featuring a 1 million‑token context window and pricing of $3 / M input tokens (95 % discount for cached) and $15 / M output tokens. It ranks #2 on the Vals…

Private certificate authority support in Amazon Bedrock AgentCore Identity
AWS ML Blog· 16 min read· Today
Private certificate authority support in Amazon Bedrock AgentCore Identity

Amazon Bedrock AgentCore Identity now supports private certificate authorities, allowing agents to authenticate to self‑hosted IdPs (e.g., Keycloak) that present TLS certificates signed by an internal CA. The PEM‑encoded root or intermediate CA is stored in AWS Secrets Manager or Amazon S3 and referenced by an…

DaisyRose-27B: Pushing the Frontier of Post-Training Efficiency
Hugging Face Blog· 5 min read· Today
DaisyRose-27B: Pushing the Frontier of Post-Training Efficiency

DaisyRose-27B, built on Qwen3.8-27B and fine‑tuned with the Miles RL framework, doubles Terminal‑Bench 4.1 pass@1 from 5.6 % to 12.0 % after 18 days of training on 144 H200 GPUs using only 1,200 curated long‑horizon tasks. The model retains a 256K‑token context window, image understanding, and the original…

Mathematical proof, powered by AI. And no hallucinations!
Amazon Science· 8 min read· Today
Mathematical proof, powered by AI. And no hallucinations!

The post announces new tools in the AutoCorrode verification suite that let AI agents interact directly with Isabelle proof assistants via the Model Context Protocol (MCP), enabling both full control (I/Q) and collaborative proof development (I/R). It also introduces Isabelle Assistant, a chat‑based interface powered…

AI’s GPU focus meets a CPU reality check at Dell and AMD
SiliconANGLE AI· Today
AI’s GPU focus meets a CPU reality check at Dell and AMD

Agentic AI is rewriting the business development playbook for chipmakers and server builders. Graphics processing units still grab the headlines, but as supercomputing and AI converge, the next wave of demand is landing on central processing units that must fit within data centers’ existing po...

How AI Agents Become a Tool for Attackers, and How to Stop It
Towards Data Science· Today
How AI Agents Become a Tool for Attackers, and How to Stop It

The article warns that AI agents can be weaponized by attackers to exfiltrate compromised credentials, infer salaries from budget data, and manipulate documents, highlighting the need for strict permission checks before any data reaches the model. It stresses that enforcing these checks at the agent layer is essential…

Neurosymbolic Agents Explained, With a Z3 Example in Python
Towards Data Science· Today
Neurosymbolic Agents Explained, With a Z3 Example in Python

The article walks through a neurosymbolic agent pipeline, illustrating how a Python‑based Z3 solver can enforce logical constraints while highlighting that the English‑to‑logic translation (via Logic‑LM, Agent‑C, and Amazon’s checks) remains unreliable. By dissecting Air Canada’s chatbot case, it shows the solver’s…

Starburst adds context support to give AI agents consistent access to business data definitions
SiliconANGLE AI· Today
Starburst adds context support to give AI agents consistent access to business data definitions

Starburst Data has extended its data‑federation platform with **Context Studio**, a new business‑context management layer that aggregates definitions from data catalogs, business‑intelligence tools, and metric stores so AI agents can retrieve consistent, trustworthy metadata when answering queries across enterprise…

Humyn Labs raises $47M to bridge the gap between AI training and real-world robotics
SiliconANGLE AI· Today
Humyn Labs raises $47M to bridge the gap between AI training and real-world robotics

Humyn Labs announced a $47 million early‑stage funding round to accelerate the deployment of intelligent robots in real‑world settings across Brazil, South Korea, and India. Accel led the $26 million equity portion, with participation from Prosus Ventures, BeeGlobal, and others. The capital will be used to bridge the…

Reducing token costs and latency in Amazon Bedrock AgentCore JSON workloads
AWS ML Blog· 14 min read· Yesterday
Reducing token costs and latency in Amazon Bedrock AgentCore JSON workloads

The team reduced token consumption by up to 71 % and cut average latency by up to 80 % when processing 20 MB+ JSON payloads in an Amazon Bedrock AgentCore production agent. By applying three composable strategies—state‑based payload partitioning, selective context injection, and caching of tool results—they kept the…

Idle, then spiking: Why AI agents are breaking Kubernetes resource planning
SiliconANGLE AI· Yesterday
Idle, then spiking: Why AI agents are breaking Kubernetes resource planning

AI agents cause workloads to remain idle for extended periods and then generate sudden resource spikes, which undermines static Kubernetes resource configurations. This unpredictable usage pattern forces platform teams to adopt autonomous orchestration to dynamically adjust resources. The shift highlights the need for…

Nvidia at SC26: Agents are the new scientists, and they need a lab
SiliconANGLE AI· Yesterday
Nvidia at SC26: Agents are the new scientists, and they need a lab

Nvidia highlighted the emerging role of AI agents as scientific collaborators at SC26, emphasizing the need for dedicated infrastructure to support their development and execution. The announcement underscores a shift from pure raw compute metrics toward agent-centric workloads in high‑performance computing…

Secure OAuth patterns for Snowflake Cortex AI through Amazon Bedrock AgentCore
AWS ML Blog· 26 min read· Yesterday
Secure OAuth patterns for Snowflake Cortex AI through Amazon Bedrock AgentCore

This article details three OAuth authentication patterns for Snowflake Cortex AI when accessed via Amazon Bedrock AgentCore, covering Snowflake‑native OAuth (per‑user Authorization Code flow), three‑legged OAuth with a corporate IdP (Authorization Code flow), and two‑legged OAuth with a third‑party IdP…

Microsoft rolls out new first-party models and updates to agentic AI continuous improvement
SiliconANGLE AI· Yesterday
Microsoft rolls out new first-party models and updates to agentic AI continuous improvement

Microsoft announced the rollout of new first‑party generative AI models and enhancements to its Foundry platform, including updates to Foundry IQ that now ship an enterprise knowledge graph out of the box to support autonomous agents. The announcement highlights tighter integration of model grounding, testing, and…

Build WhatsApp AI agents with Amazon Bedrock AgentCore
AWS ML Blog· 14 min read· Yesterday
Build WhatsApp AI agents with Amazon Bedrock AgentCore

The SMART (Social Messaging Agent Runtime Toolkit) open‑source Python SDK and Terraform stack lets you add a full WhatsApp channel to an Amazon Bedrock AgentCore agent with a single deployment command, abstracting away the integration of more than ten AWS services. It handles webhook ingestion, signature verification,…

Dell puts a single gateway between enterprise AI agents and the data they crave
SiliconANGLE AI· Yesterday
Dell puts a single gateway between enterprise AI agents and the data they crave

Dell Technologies introduced a dedicated gateway that mediates enterprise AI agents’ access to trusted, governed data, aiming to improve the value of GPU‑heavy deployments by ensuring agents operate with reliable context. The solution positions Dell as a neocloud provider enabling on‑premises enterprises to replicate…

Your complete guide to AWS Quick at AWS re:Invent 2026
AWS ML Blog· 20 min read· Yesterday
Your complete guide to AWS Quick at AWS re:Invent 2026

AWS Quick is an enterprise‑grade AI workspace announced at re:Invent 2026 that centralizes dashboards, agents, and automations for team‑wide access while enforcing tenant‑level data isolation and role‑based access controls. Customer sessions (e.g., BIZ216 and BIZ219) demonstrate how Quick’s built‑in governance…

How LoanPro built an agentic AI loan servicing assistant with Amazon Bedrock AgentCore
AWS ML Blog· 12 min read· Yesterday
How LoanPro built an agentic AI loan servicing assistant with Amazon Bedrock AgentCore

LoanPro built a multi‑agent AI loan‑servicing assistant on Amazon Bedrock AgentCore using Anthropic’s Claude Sonnet 5.5, Claude Haiku 5.5, and Claude Opus 5.5, achieving up to a 15 percent reduction in customer call times. The architecture stitches together account data from dozens of fields into a synthesized…

Responsible AI for agentic AI: A phased roadmap for platform and application teams
AWS ML Blog· 15 min read· Yesterday
Responsible AI for agentic AI: A phased roadmap for platform and application teams

This article outlines a three‑phase Responsible AI (RAI) roadmap for teams building agentic AI platforms, mapping AWS Responsible AI Lens pillars to governance foundations, pilot deployment, and evaluation/hardening stages. It highlights a real‑world case where Robinhood used a generative AI financial‑crimes agent on…

How to Improve Coding Agent Efficiency by 30% with an Orchestrator
Towards Data Science· Yesterday
How to Improve Coding Agent Efficiency by 30% with an Orchestrator

The article introduces an orchestrator pattern that adds a “herder” agent to surface pending coding questions by priority, reducing the time developers spend hunting through sessions and achieving a reported 30% efficiency gain for coding agents. By keeping the question list current via webhooks or inexpensive Haiku…

How to Govern Coding Agent Skills Across Repos With Plugins
Towards Data Science· Yesterday
How to Govern Coding Agent Skills Across Repos With Plugins

The article proposes packaging coding‑agent skills, governance rules, hooks, and Model Context Protocol (MCP) tool configurations as versioned plugins that can be distributed via a marketplace, illustrated using Claude Code and AWS. By treating these components as plugins, teams can standardize and control agent…

Ian Buck Kicks Off NVIDIA’s Week at SC26 as Science Enters the Age of Agents
NVIDIA Blog· 4 min read· Yesterday
Ian Buck Kicks Off NVIDIA’s Week at SC26 as Science Enters the Age of Agents

NVIDIA unveiled new AI agent tooling at SC26, including the Nemotron for Science model built on Nemotron 3.5 Super VL with a chemistry dataset, and showcased agents that autonomously plan experiments, run simulations, and fix software setups. Benchmarks highlighted an agent that repaired 81% of GROMACS test cases in…

Google ADK Agents That Pause, Resume, and Wait Hours for a Human
Towards Data Science· Yesterday
Google ADK Agents That Pause, Resume, and Wait Hours for a Human

Google ADK agents can now pause, resume, and wait for extended periods—up to hours—by using webhooks and HTTP 202 receipts, allowing the client connection to be released while the agent continues processing. In a demonstration, a synchronous call kept an RSS agent’s connection open for 83 seconds before the webhook…

SuperDial revamps its healthcare voice agent platform to stop AI bots wasting time calling each other
SiliconANGLE AI· Yesterday
SuperDial revamps its healthcare voice agent platform to stop AI bots wasting time calling each other

SuperDial Inc. is redesigning its healthcare voice‑agent platform with a new Agent‑to‑Agent RCM architecture to curb compute waste caused by autonomous bots calling each other. The overhaul targets inefficiencies that have emerged as AI‑driven workers proliferate in healthcare workflows. The company recently secured…

NVIDIA DGX Spark Developer Kit Turns a Desk Into a Supercomputing Lab
NVIDIA Blog· 4 min read· 2 days ago
NVIDIA DGX Spark Developer Kit Turns a Desk Into a Supercomputing Lab

NVIDIA unveiled the DGX Spark Developer Kit, a four‑node desktop‑sized HPC cluster built on the GB10 Grace Blackwell Superchip that links nodes in a ring via ConnectX‑7 networking without a switch. The kit aggregates 512 GB of memory and 20 TB of NVMe storage, enabling local deployment of up to 700 B‑parameter models…

Build memory-aware AI agents with Agentic Retrieve on Amazon Bedrock Managed Knowledge Base
AWS ML Blog· 26 min read· 2 days ago
Build memory-aware AI agents with Agentic Retrieve on Amazon Bedrock Managed Knowledge Base

Amazon Bedrock AgentCore now supports memory‑aware agents via the AgenticRetrieveStream API, letting developers add a memoryConfiguration that treats long‑term user facts as a retrieval source alongside a Managed Knowledge Base. The planner automatically decides when to pull stored user facts to enrich sub‑queries,…

How Duco transforms financial data automation with Amazon Bedrock AgentCore
AWS ML Blog· 14 min read· 2 days ago
How Duco transforms financial data automation with Amazon Bedrock AgentCore

Duco rebuilt its financial‑data automation platform on Amazon Bedrock and Bedrock AgentCore, moving from traditional ML models to managed foundation models and scalable agents. The shift enabled the processing of 20 billion transactions per month for 200+ clients—including 7 of the top 20 banks—while keeping each…

Taxonomy of multi-agent design patterns, anti-patterns, and lessons from manufacturing
AWS ML Blog· 18 min read· 2 days ago
Taxonomy of multi-agent design patterns, anti-patterns, and lessons from manufacturing

Enterprise teams building multi‑agent systems face a distributed‑systems reliability problem, with ~88 % of AI pilots never reaching production and sequential pipelines of ten 95 %‑reliable steps delivering only 59.9 % end‑to‑end reliability (twenty steps drop to 35.8 %). The article classifies coordination patterns…

How MoEngage built agentic customer engagement on Amazon Bedrock AgentCore
AWS ML Blog· 12 min read· 2 days ago
How MoEngage built agentic customer engagement on Amazon Bedrock AgentCore

MoEngage built Merlin AI, an agentic customer‑engagement platform on Amazon Bedrock AgentCore that lets marketers draft, analyze and adjust campaigns via a single conversational interface. The system uses Anthropic Claude Sonnet 4.6, 5 and 5.5 models for reasoning, tool‑call planning and content generation, and…

Novo Nordisk modernizes manufacturing process documents with Amazon Bedrock AgentCore
AWS ML Blog· 17 min read· 2 days ago
Novo Nordisk modernizes manufacturing process documents with Amazon Bedrock AgentCore

Novo Nordisk built an intelligent support system for manufacturing process documents using Amazon Bedrock AgentCore and a GraphRAG architecture that combines a knowledge graph with hybrid search. The managed agent framework coordinates retrieval and response, and a guided mode clarifies ambiguous queries, reducing…

Get answers from tables with tabular search in Amazon Bedrock Managed Knowledge Base
AWS ML Blog· 19 min read· 2 days ago
Get answers from tables with tabular search in Amazon Bedrock Managed Knowledge Base

Amazon Bedrock Managed Knowledge Base now adds tabular search, enabling agents to retrieve and compute over CSV, XLSX, and XLS files in a single request. The feature runs sandboxed code over the full table, then merges calculated results with facts from related documents to answer multi‑step analytical queries.…

NVIDIA Sweeps MLPerf Training and Endpoints as Vera Rubin Debuts With Leading Performance
NVIDIA Blog· 5 min read· 2 days ago
NVIDIA Sweeps MLPerf Training and Endpoints as Vera Rubin Debuts With Leading Performance

NVIDIA’s Vera Rubin platform swept every MLPerf Training v6.1 benchmark and, in its first MLPerf Endpoints submission, achieved more than 10× generational throughput‑per‑watt gains for agentic inference. The NVL72 system delivered up to 2× higher performance per GPU than the prior GB300 NVL72 and scaled DeepSeek‑V3…

NVIDIA CUDA-X Libraries Give Agents the Tools to Accelerate Scientific Discovery
NVIDIA Blog· 4 min read· 2 days ago
NVIDIA CUDA-X Libraries Give Agents the Tools to Accelerate Scientific Discovery

NVIDIA’s expanded CUDA‑X libraries—cuPhoton, cuPotential and cuSpa—deliver dramatic speedups for scientific workloads, cutting a month‑long black‑hole data correlation to eight minutes and boosting molecular simulation throughput by up to 3×. The cuPhoton.xBeam and cuPhoton.xLens modules run on distributed multi‑GPU…

AI Day Seoul Brings Together Korea’s Industry Leaders, Startups and Developer Communities
NVIDIA Blog· 4 min read· 3 days ago
AI Day Seoul Brings Together Korea’s Industry Leaders, Startups and Developer Communities

NVIDIA’s AI Day Seoul showcased Korean industry leaders and startups leveraging NVIDIA platforms to build physical AI and agentic systems, highlighted by Deepu Talla’s remarks on the country’s rapid robotics adoption. The event featured hands‑on workshops on deploying and optimizing AI inference at scale with the…

A Practitioner's Guide to 6 Agentic GraphRAG Patterns
Towards Data Science· 3 days ago
A Practitioner's Guide to 6 Agentic GraphRAG Patterns

This article explains how autonomous agents can transform static knowledge graphs into dynamic “Context Graphs” that retain decisions, user intent, and operational history. It outlines six distinct Agentic GraphRAG patterns that enable agents to continuously update and query graph‑based knowledge. By leveraging these…

AT&T uses agentic AI to reinvent business processes from the ground up
SiliconANGLE AI· 4 days ago
AT&T uses agentic AI to reinvent business processes from the ground up

AT&T is scaling its use of autonomous AI agents, with nearly 500 agentic workflows already in production and a goal to triple that count within a year. The company now processes roughly 45 billion AI tokens each day across its operations. This rapid expansion demonstrates how large enterprises can embed agentic AI at…

Data From 100+ AI Agents Shows They Fail by Silent Drift, Not Errors
Towards Data Science· 4 days ago
Data From 100+ AI Agents Shows They Fail by Silent Drift, Not Errors

Monitoring of AI agents should prioritize quality, cost, and latency signals, as data from over 100 agents shows that 38% of failures are first indicated by quality and token alerts, while only about 1% are flagged by explicit errors. This silent drift pattern suggests engineers need to implement proactive…

NiCE’s next-generation Cognigy bets the hard part of AI agents is running them, not building them
SiliconANGLE AI· 5 days ago
NiCE’s next-generation Cognigy bets the hard part of AI agents is running them, not building them

NiCE’s next‑generation Cognigy platform emphasizes that the hardest challenge for AI agents in contact centers is operating them at scale rather than constructing the underlying models. The announcement positions Cognigy as an execution‑focused layer that abstracts deployment, monitoring, and orchestration…

Architecture patterns for fast session startup on Amazon Bedrock AgentCore runtime
AWS ML Blog· 18 min read· 5 days ago
Architecture patterns for fast session startup on Amazon Bedrock AgentCore runtime

Amazon Bedrock AgentCore runtime V2 introduces a snapshot‑based startup that prepares the container once, freezes the microVM, and restores the snapshot for each new session, cutting cold‑start latency to roughly 2 seconds and reclaiming unused memory for cost efficiency. Engineers can achieve sub‑second latency by…

Advancing science & engineering with Antigravity’s /teamwork
Google Research· 16 min read· 5 days ago
Advancing science & engineering with Antigravity’s /teamwork

Antigravity’s /teamwork framework leverages Gemini 4 Argon, Carbon, and Hydrogen models in parallel multi‑agent workflows that pair exploratory solvers with dedicated falsifiers such as the Lean 4 proof checker and unmodified software test suites. By partitioning context per specialized agent and routing prompts…

Keyword Search Tied an 8B Embedding Model at Picking Agent Tools
Towards Data Science· 5 days ago
Keyword Search Tied an 8B Embedding Model at Picking Agent Tools

On a benchmark of 200 ToolRet questions covering up to 3,794 tools, BM25 keyword search ran roughly 40× faster than the 8 B BGE‑large embedding model on a CPU, and using full tool documentation made BM25 the top‑performing configuration. This shows that a classic lexical retriever can outpace a large embedding model…

Agents idling while they wait on humans push Kubernetes to hand back unused resources
SiliconANGLE AI· 5 days ago
Agents idling while they wait on humans push Kubernetes to hand back unused resources

Enterprises are observing AI agents idling while awaiting human input, prompting Kubernetes to reclaim the now‑unused compute resources. This shift highlights Kubernetes as a critical component for managing autonomous software workloads rather than just a static container orchestrator. Engineers must consider dynamic…

What AI Agents Speed Up in PhD Research, and What They Can't
Towards Data Science· 5 days ago
What AI Agents Speed Up in PhD Research, and What They Can't

The article reflects on how AI agents can accelerate routine tasks such as faster code execution, experiment iteration, and draft writing for a PhD student in machine learning and ecology, but they cannot resolve underlying data quality issues that require domain insight. It outlines three predictions about how…

AI agent tooling provider LangChain secures $300M investment
SiliconANGLE AI· 6 days ago
AI agent tooling provider LangChain secures $300M investment

LangChain Inc., the open‑source AI agent tooling startup, announced a $300 million financing round led by Alphabet’s CapitalG, valuing the company at $3 billion. The round also included investors IVP, Sapphire Ventures, GIC and DFJ Growth. This sizable capital injection underscores strong market confidence in…

As AI agents take control, Nvidia rethinks the security playbook
SiliconANGLE AI· 6 days ago
As AI agents take control, Nvidia rethinks the security playbook

Nvidia introduced **Scale‑In**, a new pillar in its AI networking stack that provides purpose‑built acceleration for securing, managing, and operating “agentic AI factories.” The offering focuses on secure access, high‑performance data movement, tenant isolation, simpler operations, and more predictable performance as…

How Natera designed an agentic AI system for lab ops with Amazon Bedrock AgentCore
AWS ML Blog· 13 min read· 6 days ago
How Natera designed an agentic AI system for lab ops with Amazon Bedrock AgentCore

Natera built SWIFT (Systematic Workflow Analytics and Resolutions Manager) on Amazon Bedrock AgentCore to automate lab‑instrument support tickets, achieving ≈130× faster first‑response times and 96.2 % automated coverage. The system leverages two agents sharing a single skill library, integrates tightly with Jira,…

Kubernetes gets a second act as AI agents turn one prompt into thousands of inference calls
SiliconANGLE AI· 6 days ago
Kubernetes gets a second act as AI agents turn one prompt into thousands of inference calls

Kubernetes is being repurposed to orchestrate the massive number of inference calls generated by agentic AI systems, where a single user prompt can trigger thousands of downstream model invocations. The platform’s reliability and predictability are leveraged to manage this surge as AI spending shifts from model…

Google retools Kubernetes for AI agents instead of building a new stack
SiliconANGLE AI· 6 days ago
Google retools Kubernetes for AI agents instead of building a new stack

Google is redesigning Kubernetes to better support AI agent workloads, addressing challenges like untrusted code generation, long idle periods, and rapid pod churn that stress the control plane. The retooling focuses on adapting the existing orchestration platform rather than creating a new stack, aiming to improve…

The Pulse: Details of incident.io’s AI spend
Pragmatic Engineer· 9 min read· 6 days ago
The Pulse: Details of incident.io’s AI spend

incident.io reports that AI agents now generate code for nearly 80% of production pull requests, driving a 6× rise in “effort points” and a 60% drop in incidents per 1,000 PRs. The company’s AI spend has climbed to roughly $1.6 M this year (~$2,700 per engineer per month), while release velocity rose 47% to 2.8…

Introducing Gemini 4 Hydrogen
Google DeepMind· 9 min read· 6 days ago
Introducing Gemini 4 Hydrogen

Gemini 4 Hydrogen is a new 1 M‑token context LLM priced at $1.50 / M input and $7.50 / M output tokens, achieving an Artificial Analysis Intelligence Index score of 51 (up to 54 at xhigh) and delivering 2.5× faster execution than Gemini 3.8 Flash while tripling Terminal‑Bench 4.1 scores to 57%. The Orchestrator…

Run agents securely and efficiently with the next generation AgentCore runtime
AWS ML Blog· 9 min read· 6 days ago
Run agents securely and efficiently with the next generation AgentCore runtime

Amazon Bedrock AgentCore’s new runtime delivers hardware‑isolated Firecracker MicroVM sessions that start up to 15× faster, can scale to 4× more compute, and run up to 68 % more efficiently while preserving isolation. The updates let agents dynamically increase memory and CPU within each session, supporting heavier…

Migrate to Amazon Bedrock Managed Knowledge Base with your coding agent
AWS ML Blog· 19 min read· 6 days ago
Migrate to Amazon Bedrock Managed Knowledge Base with your coding agent

The article announces a migration guide in the AWS Agent Toolkit that lets a coding agent help move a customer‑managed Amazon Bedrock knowledge base to the new Amazon Bedrock Managed Knowledge Base service. Managed Knowledge Base offloads vector store provisioning, embedding model tuning, hybrid search, reranking and…

Build composable, interoperable multi-agent systems with A2A on AWS
AWS ML Blog· 26 min read· 6 days ago
Build composable, interoperable multi-agent systems with A2A on AWS

The article introduces the Agent-to-Agent (A2A) protocol, enabling secure, bidirectional communication between a LangChain orchestrator running on Amazon ECS and a Strands agent hosted on Amazon Bedrock AgentCore. By leveraging OAuth‑authenticated calls, the system allows agents to exchange data across organizational…

Fine-grained access control for Amazon Bedrock Managed Knowledge Base tools
AWS ML Blog· 24 min read· 6 days ago
Fine-grained access control for Amazon Bedrock Managed Knowledge Base tools

This article presents a pattern for enforcing fine‑grained access control in Amazon Bedrock agents by injecting verified user attributes into Model Context Protocol (MCP) tool calls via the AgentCore Gateway, then leveraging Managed Knowledge Base’s ACL‑aware search that filters results based on the userContext…

From prototype to production: Engineering agents with Agent DLC
AWS ML Blog· 15 min read· 6 days ago
From prototype to production: Engineering agents with Agent DLC

The article introduces the Agent Development Lifecycle (Agent DLC), a framework for moving AI agents from prototype to production by adding systematic checks for non‑determinism, prompt/tool changes, and external dependency drift. It mandates running each test case k times to evaluate pass‑rate distributions,…

How Block powers collaborative AI agents with Buzz and Unity Gateway
Databricks Blog· 6 min read· 6 days ago
How Block powers collaborative AI agents with Buzz and Unity Gateway

Block introduces Buzz, a platform that unites employees and AI agents in a shared workspace, leveraging Unity Gateway for collaborative interactions. The announcement highlights integration with Databricks to facilitate data‑driven agent workflows. This architecture aims to streamline multi‑agent coordination for…

Build autonomous AI agents with Amazon Bedrock and Spring AI
AWS ML Blog· 21 min read· Nov 12, 2026
Build autonomous AI agents with Amazon Bedrock and Spring AI

Amazon Bedrock AgentCore combined with Spring AI lets Java teams build autonomous AI agents that can plan, invoke tools, and retain memory across multi‑turn conversations, all using familiar Spring patterns and managed runtime infrastructure. The walkthrough demonstrates a single‑agent travel assistant (tool calling +…

Introducing the Web Search API on Amazon Bedrock AgentCore
AWS ML Blog· 15 min read· Nov 11, 2026
Introducing the Web Search API on Amazon Bedrock AgentCore

Amazon Bedrock AgentCore now offers a standalone Web Search API called **InvokeWebSearch**, letting developers retrieve ranked web results—including text extracts, titles, URLs, and publication times—with a single SDK or CLI call. The request is signed with AWS Signature Version 4 using existing IAM credentials,…

ICYMI: What landed for AI builders in October 2026
AWS ML Blog· 8 min read· Nov 11, 2026
ICYMI: What landed for AI builders in October 2026

AWS announced October 2026 updates that add new high‑throughput, low‑cost models to Amazon Bedrock and extend the AgentCore platform, letting builders match tasks to the optimal model, keep agents running when primary models are throttled, and gain finer cost predictability. Claude Haiku 5.5 offers up to 75 % lower…

Using wearables to enable cardiovascular-kidney-metabolic (CKM) risk profiling for personalized coaching
Google Research· 9 min read· Nov 11, 2026
Using wearables to enable cardiovascular-kidney-metabolic (CKM) risk profiling for personalized coaching

The paper presents an investigational machine learning framework that converts passive wearable data into a modifiable “coachable health age” and uses a sensor‑world model (SensorWM) trained on over 2.2 million person‑weeks to simulate personalized 12‑week “what‑if” habit trajectories, which are then turned into…

EXPLORE AI NEWS

Daily hand-picked stories on LLMs, RAG, agents and production AI — curated for engineers who ship.

BROWSE NEWS

GET THE WEEKLY DIGEST

Join engineers getting the Monday signal-over-noise AI breakdown. No spam, unsubscribe anytime.

LEARN AI ENGINEERING

Curated courses, research papers, repos and tutorials built for engineers leveling up in AI.

START LEARNING