All posts tagged: Agents

AI agents are quietly generating chaos engineering failures enterprises don’t track yet

AI agents are quietly generating chaos engineering failures enterprises don’t track yet

Published by skeptic

There is a category of production incident that engineering teams are not tracking yet — because it doesn’t fit any existing postmortem template. The agent initiated an action. The action was technically correct given the agent’s context. The context was incomplete. The infrastructure cascaded. And, by the time the incident review happened, three teams were arguing about whether it was an agent failure or an infrastructure failure,  because the frameworks for thinking about these two things have never been connected. The scale of this exposure is no longer theoretical. Seventy-nine percent of organizations now have some form of AI agent in production, with 96% planning expansion. Gartner predicts 33% of enterprise software will include agentic AI by 2028, but separately warns that 40% of those projects will be canceled due to poor risk controls. What neither statistic captures is the failure mode happening between those two numbers: Agents that are running, that are not canceled, and that are quietly generating infrastructure events no one has categorized as risk. I’ve spent six years building infrastructure automation …

Why have immigration agents detained this American citizen three times?

Why have immigration agents detained this American citizen three times?

Published by Sam Flores

When immigration agents pulled U.S. citizen Leonardo Garcia Venegas from his car this month and shackled him, he wasn’t surprised. He wasn’t scared. He was tired. As ProPublica detailed last fall, he had already been detained twice before. A year ago, Garcia Venegas was filming his brother’s arrest during a raid on their coastal Alabama construction site when he was tackled by agents, who ignored his pleas that he was a citizen. A few weeks later, an officer entered the home Garcia Venegas was building and refused to trust the now-26-year-old’s Alabama REAL ID, which only citizens and legal residents can get. Videos of the incidents went viral. He appeared before Congress. He also has a suit pending against the Trump administration. But all the attention hasn’t changed much. On May 2, agents followed him back to his home. They again didn’t believe his claims of citizenship or the REAL ID he once again tried to show them. Now, after that latest detention, Garcia Venegas sounds demoralized. “Honestly, it feels terrible,”  Garcia Venegas told ProPublica. …

Gunman shot dead by Secret Service agents near White House: What we know | News

Gunman shot dead by Secret Service agents near White House: What we know | News

Published by Sam Flores

EXPLAINER No Secret Service agents sustain injuries, but a bystander is wounded in the exchange of fire. A man has been shot dead by United States Secret Service officers after opening fire on a security checkpoint near the White House, and a bystander has been wounded in the gunfire. Several US media outlets have identified the gunman as Nasire Best, a 21-year-old man from the neighbouring state of Maryland who was known to the Secret Service and had a documented history of mental health conditions. Recommended Stories list of 3 itemsend of list Shortly after 6pm (22:00 GMT) on Saturday, the suspect approached a Secret Service checkpoint at the intersection of 17th Street and Pennsylvania Avenue in Washington, DC, pulled a weapon from his bag and began shooting at officers posted there, the Secret Service said in a statement. The gunman was taken to hospital where he was pronounced dead. No Secret Service officers were wounded. President Donald Trump was in the White House during the incident but “no protectees or operations were impacted,” the …

Setting Up AI Voice Agents for Business and Personal Use in 2026

Setting Up AI Voice Agents for Business and Personal Use in 2026

Published by skeptic

Integrating a phone number with Hermes Agent, powered by Vapi, allows for AI-driven automation of phone-based tasks. By using features like real-time transcription, voice synthesis, and external integrations, the system can handle scenarios such as confirming appointments, managing inquiries, or conducting lead follow-ups. According to David Ondrej, users can configure Hermes Agent to autonomously make outbound calls, offering a structured approach to managing communication workflows. Learn how to set up Hermes Agent on different platforms, such as local environments or VPS hosting and configure API access for operational readiness. Explore how Vapi supports call management through detailed logs and customizable voice interactions. Gain insight into advanced customization options, including adapting voice models and designing agents for specific use cases. How to Set Up Hermes Agent TL;DR Key Takeaways : Hermes Agent, powered by Vapi, automates phone-based tasks such as outbound calls, inbound inquiries, appointment scheduling and lead generation using AI-driven features like real-time transcription and voice synthesis. The system offers a flexible setup process, accommodating users with varying technical expertise and can be deployed locally …

Your AI agents need a terminal, not just a vector database

Your AI agents need a terminal, not just a vector database

Published by skeptic

When agentic workflows fail, developers often assume the problem lies in the underlying model’s reasoning abilities. In reality, the limited information provided by the retrieval interface is often the primary limiting factor. Researchers at multiple universities propose a technique called direct corpus interaction (DCI) that lets agents bypass embedding models entirely, searching raw corpora directly using standard command-line tools. The limits of classic retrieval In classic retrieval systems such as RAG, documents are chunked, converted into vector representations (or embeddings), and indexed offline in a vector database. When an AI system processes a query, a retriever filters the entire database to return a ranked “top-k” list of document snippets that match the query. All evidence must pass through this scoring mechanism before any downstream reasoning occurs. But modern agentic applications demand much more. “Dense retrieval is very useful for broad semantic recall, but when an agent has to solve a multi-step task, it often needs to search for exact strings, numbers, versions, error codes, file paths, or sparse combinations of clues,” the authors of the …

D&B’s database of 642 million businesses was built for humans, not AI agents. So they rebuilt it.

D&B’s database of 642 million businesses was built for humans, not AI agents. So they rebuilt it.

Published by skeptic

Dun & Bradstreet has spent over 180 years building a comprehensive commercial database. Its Commercial Graph, covering 642 million businesses and their relationships, corporate hierarchies and risk profiles, was designed for people. Credit analysts, risk managers and sales professionals who could wait for query results and work through ambiguous entity matches. AI agents cannot do any of those things. When D&B’s customers started pushing agents into credit, procurement and supply chain workflows, the Commercial Graph that had reliably served nearly 200,000 customers globally became a problem. The systems built to serve human analysts were the wrong architecture for machines. So D&B rebuilt. “We need to think about agents as our new consumer category, evolving from our standard credit analysts or sales and marketing professionals, et cetera, to also now catering to these customers’ agents,” Gary Kotovets, Chief Data and Analytics Officer at Dun & Bradstreet, told VentureBeat. What broke when agents started querying The Commercial Graph was not a single database. It was a collection of separate systems built for different use cases and different …

Eufy Brings Local AI Agents to Home Security, the First I’ve Seen Without a Subscription

Eufy Brings Local AI Agents to Home Security, the First I’ve Seen Without a Subscription

Published by skeptic

Eufy is one of my favorite choices for no-subscription home security cameras, so I was interested in what the brand would unveil for Anker Day on Thursday. One announcement caught my eye immediately: Eufy is releasing a new AI agent called EdgeAgent that will exist locally on a new companion device called the Smart Security Shield. Smart Security Shield is a sensor and camera with a 180-degree field of view and equipped with digital security key technology to prevent unauthorized access. It’s primarily designed for outdoor use, where the onboard AI chipset uses facial recognition to identify friends and family from up to 100 feet away. Eufy says the device will work with multiple hardware and service bundles from the company. Because EdgeAgent is entirely local, it doesn’t need to exchange data with online servers. Eufy says that makes the AI processing 63% faster than competitors’ cloud-based systems, cutting delays to only a few seconds. EdgeAgent will be able to process data and activate specific security measures based on its analysis, such as turning on a spotlight, …

A 0.12% parameter add-on gives AI agents the working memory RAG can’t

A 0.12% parameter add-on gives AI agents the working memory RAG can’t

Published by skeptic

AI agents forget. Every time a coding assistant loses track of a debugging thread, or a data analysis agent re-ingests the same context it already processed, the team pays in latency, token costs, and brittle workflows. The fix most teams reach for — expanding the context window or adding more RAG — is increasingly expensive and still doesn’t reliably work. To address this, researchers from Mind Lab and several universities proposed delta-mem, an efficient technique that compresses the model’s historical information into a dynamically updated matrix without changing the model itself. The resulting module adds just 0.12% of the backbone model’s parameters — compared to 76.40% for one leading alternative — while outperforming it on memory-heavy benchmarks. Delta-mem allows models to continuously accumulate and reuse historical data, reducing the reliance on massive context windows or complex external retrieval modules for behavioral continuity. The long memory challenge The conventional solution is to simply dump all the information into the model’s context window. But as Jingdi Lei, co-author of the paper, told VentureBeat, current systems treat memory …

Google’s Managed Agents API promises one-call deployment at the cost of execution layer control

Google’s Managed Agents API promises one-call deployment at the cost of execution layer control

Published by skeptic

At Google I/O, the company unveiled Managed Agents in its Gemini API — a service that promises to collapse weeks of agent deployment work into a single API call. It’s also a sign that Google believes its ecosystem, including the newly launched Antigravity CLI, is ready to own the execution layer end-to-end. Before a single agent is written, teams are already spending days on the unglamorous work: standing up execution environments, managing sandboxes, wiring tool call infrastructure. Model providers like Anthropic have launched platforms to handle much of that work — but Google’s approach is different. Google said in a blog post that Managed Agents in the Gemini API abstracts “away the complexity so that you can focus on your product experience and agent behavior.” The service is available in preview via new custom templates in Google AI Studio. The growth has introduced a real architectural question: should agent management live at the execution layer — embedded in the model or its harness — or at the infrastructure layer, as a separate runtime? Comparing Google’s …

Enterprise AI agents keep failing because they forget what they learned

Enterprise AI agents keep failing because they forget what they learned

Published by skeptic

RAG architectures are good at one thing: surfacing semantically relevant documents. That’s also where they stop. A framework called a decision context graph addresses that gap by giving agents structured memory, time-aware reasoning, and explicit decision logic. Rippletide, a startup in the Neo4j ecosystem, has built one. The key capability: agents that are non-regressive, able to freeze validated sequences of actions and compound on them over time. “The key point you want is non-regressivity: How do you make sure that, when the agent will generate something new, you can compound on the previous discoveries?” said Yann Bilien, Rippletid’s co-founder and chief scientific officer.  Why RAG doesn’t go far enough Enterprise context is sprawled across ERP tools, logs, databases, vector stores, and policy documents. Generative AI tools can retrieve from all of it — through keyword search, SQL queries, or full RAG pipelines — but retrieval has a ceiling. Notably, data retrieved may not be relevant to the decision at hand (thus causing hallucinations); and, even if agents do pull the right data, they often lack …