Part 2 of The Intellyx Building Agentic Resilience Series by Eric Newcomer, for Gremlin The identifies the core issue: generating application code using AI tools is not delivering promised productivity gains, mainly due to unresolved issues such as testing, and ensuring AI agents behave as expected. One of the biggest challenges in deploying and governing AI generated code is discovering and preventing incidents and outages resulting from agentic hallucinations and errors.
Architect, analyst, fractional CTO. Database, transaction processing, cloud, distrubuted systems. Also craft beer, blues, and jazz.
Is this you? As a journalist, you can create a free Muck Rack account to customize your profile, list your contact preferences, and upload a portfolio of your best work.
Claim your profile
Articles by Eric Newcomer
WSO2 Rebrands as Choreo Labs, Pivots to AI Governance
An Intellyx Brain Candy Update When we last wrote about WSO2, they were focused on developing open source products for API management, identity access management, and on developing a SaaS based internal developer platform (IDP) called Choreo. They have since open sourced Choreo and submitted it to the Cloud Native Computing Foundation (CNCF) as a sandbox project called Open Choreo.
The Many Facets of Sovereign Infrastructure
Continuously evolving regulations, competitive pressures, and the increasing use of generative AI makes protecting and controlling access to computing resources more important than ever. Private clouds are an attractive way to achieve data sovereignty and guard against data leakage, but it isn’t as simple as implementing data protections. Many additional facets have to be taken into account for any regulatory compliant design and deployment.
To make agentic AI adoption safer, root out the rogue agents
Part 4 of the Intellyx Agentic SecOps Series for Straiker, by Eric Newcomer In this series, we started at the beginning point of agentic security, where attacking agents exploit production agents, explored how red, blue, and purple security teams are adopting agents, and stepped into an Agentic SecOps Command Center to align teams and better cover this rapidly expanding attack surface.
NeverBlink: AI Assisted DBA for SRE and/or Platform Engineering
An Intellyx Brain Candy Brief NeverBlink offers a specialized AI-driven platform for database maintenance and optimization, supporting PostgreSQL, ClickHouse, and OpenSearch. The platform offers Platform Engineers and SREs deep, database-specific intelligence that generic AI tools lack. In particular, NeverBlink monitors query optimization and calculates a database “health score” based on stability, performance, resilience, and security measurements.
Continuent: Auree for Automated HA Database Deployment
An Intellyx Brain Candy Update When we last covered Continuent, they described their highly available Tungsten Clusters for MySQL and MariaDB databases. In our recent briefing, Continuent talked about their new Auree product, layered on the Tungsten Cloud and developed by the same team, which automates database high availability (HA) and controls deployment inside your cloud account or across multiple cloud accounts. Auree also supports full data sovereignty.
Sevii: Agentic AI for Autonomous Cyber Defense
An Intellyx Brain Candy Brief Sevii’s agentic AI “warriors” remediate cyberattacks at machine speed and scale without requiring humans in the loop. Cybercriminals use AI to rapidly attack organizations at increasing volumes. In this modern escalation of the conflict between cybercriminals and cyberdefenders, Sevii’s security framework fights agentic fire with agentic fire, automating security analyst workflows to keep pace.
SecOps Agents Are Red and Blue, and Can be Purple, Too
Article,BrainBlog,Featured,SD Times Article | Eric Newcomer BrainBlog in ITOpsTimes, for Straiker Regardless of what the calendar says, 2026 is the year of the AI agent. They are appearing all over the IT landscape, introducing new threats and vulnerabilities as they do. Organizations rapidly build and deploy AI agents with AI driven coding tools, continuously increasing the number of deployed agents with little or no oversight and governance.
NuoData: Anti Gravity Data Governance for AI
An Intellyx Brain Candy Brief NuoData inverts the “center of gravity” approach to AI data governance. Instead of moving data into a data lake such as Snowflake, Databricks, or Fabric to use it for AI, NuoData governs an organization’s data in motion. In NuoData’s approach, governance, security, and access policies apply to data in flight rather than after the data reaches a consolidated destination.
Prevalent AI: Curated Knowledge Graph for AI Cybersecurity Defense
An Intellyx Brain Candy Brief Coming out of the GCHQ (the UK’s NSA equivalent), the founders of Prevalent AI have a lot of experience handling large scale cybersecurity threats with the potential for significant impact. Gen AI and agentic AI cybersecurity threats are current examples. Prevalent AI’s entry into the AI cybersecurity market adapts their proven technology to respond to the growing incidence of AI generated threats.
SecOps Agents Are Red and Blue, and Can be Purple, Too
Regardless of what the calendar says, 2026 is the year of the AI agent. They are appearing all over the IT landscape, introducing new threats and vulnerabilities as they do. Organizations rapidly build and deploy AI agents with AI-driven coding tools, continuously increasing the number of deployed agents with little or no oversight and governance.
Start your AI agent testing with deterministic tooling
By now everyone is aware of the limitations inherent in generative AI and the AI agents that use it to complete their tasks, and the challenges involved in getting them enterprise quality. If you are planning to incorporate AI agents into your enterprise IT architecture, how are you planning to validate their quality and accuracy? Perhaps you are thinking about using testing agents to ensure your enterprise agents produce correct results?
The Agentic AI Transformation of DevSecOps
Conventional wisdom says that adopting gen AI requires you to re-think and re-examine how your organization functions. This is true for any new technology, especially for one as revolutionary as gen AI. But first you need a clear understanding of what the new technology is good for, what it does, and how it works. When AI agents do a lot of the manual work people used to do, organizational structures change to reflect that.
Zombie Workloads Haunt Data Center Efficiency Efforts
Blog Post Mention,Featured | Cloud computing pioneered the use of commodity hardware, massive scale, aggressive replication, and container orchestration tools like Kubernetes. But these innovations don’t easily map to today’s LLMs and agentic AI. The GPU-based hardware behind modern AI is far different from the commodity setups built for cloud-style distributed computing – and AI zombies cost more, in both money and electricity.
Coro: Cybersecurity Workspace Platform for SMBs
An Intellyx Brain Candy Brief Coro offers an all-in-one cybersecurity platform for small to medium businesses (SMBs) and small IT organizations. Such a platform is also called a cybersecurity workspace. The platform provides endpoint security, email security, data loss protection (DLP), compliance auditing, security awareness training, and network security with integrated monitoring and reporting tools.
Spegel: Reducing Image Loading Latency for Kubernetes
An Intellyx Brain Candy Brief Spegel is an open source project that reduces latency for loading images into a Kubernetes cluster. Instead of going to the image registry (OCI) each time a copy of an image is requested, Spegel retrieves the image from a peer node. Peer-to-peer image distribution turns every warm node in a Kubernetes cluster into a registry mirror, so the upstream registry stops being a single point of contention.
Haven from MirrorTab: Helping you Safely Navigate the Internet
An Intellyx Brain Candy Brief As my colleague Jason English noted in his last year’s SiliconAngle article about RSAC 2025, MirrorTab blocks in-browser attacks by executing HTML and JavaScript on a protected server instead of in the browser, streaming the resulting pixels to the user’s session. This approach presents malicious code from taking over a browser session.
How to kick the hyperconverged infrastructure habit
Infrastructure convergence is constantly trending toward improved abstraction, operation, and control. Similar to the way a traditional operating system abstracts, controls, allocates, and monitors compute, storage, and network resources for monolithic applications, converged infrastructure operating systems are emerging to meet the same requirements for highly distributed, cloud native applications.
Application Modernization is not Just Pushing the AI Button
Intellyx Cortex Newsletter by Eric Newcomer, Principal Analyst Application modernization is a top priority for many organizations, and the market is growing at 15% a year. Much of the demand is driven by the need to ensure systems are AI-enabled. Gartner released the AI Augmented Code Modernization Magic Quadrant August 3, which shows Microsoft, AWS, and Moderne as the leaders, with Rocket Software and BMC as the challengers.
Dawnguard: Cloud Native Security with AI Enhanced Design
An Intellyx Brain Candy Brief Designing secure cloud native infrastructure involves coordination across multiple stakeholders, including architects, developers, cybersecurity experts, and business product representatives. It’s not always easy to achieve consensus across stakeholders with different roles and reponsiblities, but having a common diagram to center the discussion around really helps. Creating such diagrams requires art as well as science.
Calero: Managing Enterprise Tech Spend
An Intellyx Brain Candy Brief It’s difficult to collect and present an organization wide total for technology spend, along with the details that give meaning to the spend. It’s also hard to accurately and completely define technology spend and what should be included and how. And it’s even harder to figure out whether the technology is delivering appropriate value for the spend.
Gimlet Labs: Multi-Silicon Neocloud for Efficient Inference Processing
An Intellyx Brain Candy Brief As AI adoption increases, so do token charges. New models basically say to us “AI does more, but it costs more.” Therefore, it seems like token costs are going to continue to rise, and may even constrain AI adoption and use. The good news is that the generative AI industry is starting to focus more now on reducing token costs and increasing the efficiency of processing AI workloads.
DataHub: Metadata Based Context Graph for Analytics Agents
An Intellyx Brain Candy Brief It’s difficult for any enterprise to get a handle on the entirety of its data landscape. Data models typically support individual applications, not enterprise wide characteristics such as total revenue, expense, or profit. DataHub started life at LinkedIn as a solution to this problem, collecting organizational metadata into a common catalog and through that aggregation offering a common place to understand and govern a data landscape.
Fleet Device Management: Flexible, Customizable, and Autonomous
An Intellyx Brain Candy Brief Endpoint management tools for enterprise devices tend to view the world of end user devices as another task for the server admin, pushing software and patches to PCs and mobile devices the same way they push software and updates to servers.
Atsign: Built In Security for Agentic Communication
An Intellyx Brain Candy Brief Network security is often an afterthought, especially when using AI generated code to develop applications and agents. Atsign builds in network level security as you generate code, using an MCP server to access a security blueprint for your application or agentic network communication pattern.
Data Quality for Gen AI: Can MCP server integrations replace data lakes?
Traditional data lake architecture It almost goes without saying by now, although many people will still say it, perhaps because it remains a challenge for many organizations, that: The quality of the results from using AI applications depends almost entirely on the quality of the data available to them. Organizations typically build and fill a data lake to address data quality issues, and in doing so “clean” and standardize data from multiple sources.
Using AI for legacy data modernization? Don’t fall into this trap
If the future of applications is in genAI, then the future of data for AI is in the cloud. The effectiveness of genAI applications depends almost entirely on the quality of data those applications work with. The goal of any legacy data modernization project includes laying the foundation for AI applications.
MotherDuck: Cloud Data Warehouse with MCP Server for Duck DB
An Intellyx Brain Candy Brief MotherDuck is a data warehouse utility built using a shared, serverless cloud deployment of DuckDB. DuckDB is an in-process embedded database system that runs in memory for high performance access to large amounts of data. MotherDuck’s data warehouse implementation of DuckDB integrates with a wide variety of tools for data ingestion, querying, and application development.
Dispersive: Secure, Resilient Networking at Scale
An Intellyx Brain Candy Brief Security attacks increase dramatically every year. Generative AI accelerates the pace of attack as well as the risk of data breaches, fraud, and IT outages. Effectively operating as pubic service, the Internet broadly exposes a significant number of vulnerable attack points and intermittent connectivity challenges.
Modernizing telecom billing systems with real-time event streaming
Every telecom bill starts with recording an event – a phone call, text message, or stream of data. Each call detail record (CDR), usage detail record (URD), or Internet protocol detail record (IPDR) eventually produces a consumption charge on a telecom bill. How and when a telecom billing system processes each network consumption event and creates bills has a significant impact on cost and customer service.
Datadog and T-Mobile leaders reveal the reality of deploying AI agents in production
Article,Featured,The New Stack Article | Article in The New Stack by Eric Newcomer With careful governance and validation, AI agents are increasingly gaining adoption for specific enterprise functions, according to speakers at the AI Agent Conference this week in New York. The popularity of new, highly capable AI coding agents has grown dramatically over the past year.
AI startups are scrambling to survive in big tech’s shadow
Article,Featured,The New Stack Article | Article in The New Stack by Eric Newcomer Startups in the AI world gathered at the AI Agent Conference this week in New York are scrambling to carve out a niche to avoid being overshadowed by industry leaders. “And it’s not even just startups, but for example what Claude has done to Figma and Canva,” Omer Trajman, founder of AskFora, told The New Stack during the conference, held at a hotel in Midtown Manhattan.
New Agentic AI Tool Analyzes Oracle Fusion and Workday Releases
LerenLeren Article about Opkey’s Release Advisor by Agus Wijaya [excerpt] The launch of Release Advisor has garnered attention from industry experts who recognize the pressing need for such a solution. David Zimmerman, VP of Product and Solutions at Opkey, articulated the strategic impetus behind the product: “Enterprise software innovation is accelerating rapidly, but many organizations struggle to keep up with the pace of change.
Datadog and T-Mobile leaders reveal the reality of deploying AI agents in production
With careful governance and validation, AI agents are increasingly gaining adoption for specific enterprise functions, according to speakers at the AI Agent Conference this week in New York. The popularity of new, highly capable AI coding agents has grown dramatically over the past year. Still, the code they generate cannot be trusted in production, said Datadog’s Chief Scientist, Ameet Talwalkar, in the opening keynote.
From Walking to Running the Marathon: Building Your Agentic AI Roadmap
An old Chinese proverb says that every journey of a thousand miles starts with a single step. When starting out, adopting agentic AI can seem like a journey of a thousand miles, at least as you try to keep up with the latest and greatest industry innovations. Some agentic AI promoters go to the extreme, asking you to believe that AI agents will replace human employees and act without supervision to carry out any task you can think of.
Embrace: Observing the User Experience Impact
An Intellyx Brain Candy Brief Mobile application users can be extremely sensitive to application usability and latency issues. If an application is not intuitive enough, reliable enough, or doesn’t respond quickly enough, users tend to switch to another app. During the early days of mobile applications, organizations realized they needed to capture and analyze user experience data to continuously improve their applications and remain competitive.
Appdome’s IDAnchor™ Gets New Risk Intelligence APIs
Appdome’s IDAnchor™ Gets New Risk Intelligence APIs to Power Mobile Reputation & the Enterprise AI Age “Mobile devices are more a part of daily life than ever,” said Eric Newcomer, Principal Analyst at Intellyx. “Cybercriminals and fraudsters know this and are constantly ramping up their attacks. Fortunately, mobile defense vendors such as Appdome continue to respond.
FusionAuth: Developer Oriented Authentication and Authorization Controls
An Intellyx Brain Candy Brief In the modern cybercrime environment you cannot simply deploy an application into production on the network without basic security controls that prevent unauthorized access. These controls start with identity (ID) management, that is, by creating a username and password (or alternate method of logging on for an ID) and controlling access control for those IDs once logged on.
Pomerium: Network Based Context Authorization
An Intellyx Brain Candy Brief Authorization is essentially an unsolved problem in computing. Partial solutions exist, but it’s still not possible to define in any standard way exactly which data items in a database which people can access or update, for example. Coarse grained solutions are on the market, but they tend to start with the server side view – in other words, defining a role based access control matrix or a set of permissions associated with a resource such as a file, database, or API.
emma: Abstracting and Connecting Infrastructure Across Clouds and On-Prem
At first it was all about moving workloads to AWS. Then it was about multi-cloud deployment choices across Azure, Google Cloud, IBM, and Oracle. Then it was all about repatriation and private clouds and moving workloads back to on-prem. All of this movement happens for good reasons, but moving workloads is not always easy and can be costly. Many organizations choose multiple options, though, and deploy workloads in multiple public clouds as well as on-prem. This is where emma technologies comes in.
MetalBear: Isolated Virtual Testing in a Staging Enironment
An Intellyx Brain Candy Brief Open source project mirrorD, developed by MetalBear, creates a virtual space within an organization’s staging environment to run locall processes in their Kubernetes environment for testing microservices and AI agents. Staging environments are typically used only as a pre-production environment to test the application deployment process. mirrorD enables customers to further utilize a staging environment for microservice and agent testing.
LocalStack: Local Sandbox for Testing Cloud Applications
An Intellyx Brain Candy Brief LocalStack‘s product provisions a local dev and test environment for AWS and Snowflake cloud based applications. Developers can test their applications locally on their PCs, laptops, or local servers before deploying the applications to production in the cloud. The LocalStack AWS environment replicates production environment APIs on local devices, allowing you to test against them locally.
The Future of Data Experience Management for Security and Observability
An Intellyx BrainBlog by Eric Newcomer | Part 4 of the Auguria Data Experience Series Security and observability data present different views of incidents and outages from the application and network sides. In a pointillist painting, the artist’s careful placement of individual colored dots creates a coherent picture seen from a distance, which is one way to think about the potential result of combining these two different sources of data.
Will Kubernetes Become the Standard for AI Workloads?
Article,Featured,SD Times Article | Article for SD Times by Eric Newcomer The answer, of course, is that it depends. But the Cloud Native Computing Foundation (CNCF) announced some major steps forward toward this goal at the KubeCon Europe conference last week in Amsterdam. This was the biggest KubeCon ever with about 13,500 attendees, representing an 8% growth over last year, reflecting CNCF’s extraordinary success in establishing Kubernetes as the standard for container orchestration.
MCP maintainers from Anthropic, AWS, Microsoft, and OpenAI lay out enterprise security roadmap at Dev Summit
Article,Featured,The New Stack Article | Article in The New Stack by Eric Newcomer In a roundtable panel at the MCP Dev Summit last week in New York, Model Context Protocol (MCP) maintainers from Anthropic, AWS, Microsoft, and OpenAI reassured us that the MCP spec is in safe hands at the Agentic AI Foundation (AAIF) and will be addressing critical enterprise requirements for security, reliability, and governance.
Akamas: Analyzing Observability Data for Continuous Configuration Improvement
An Intellyx Brain Candy Brief Akamas continuously analyzes observability data to recommend and generate application configuration improvements, on Kubernetes or not, using AI to analyze metrics data and more efficiently allocate application resources. Akamas ingests data from observability tools using APIs and continuously analyzes ways to allocate resources to improve application performance.
OpenObserve: High Performance, High Efficiency, Open Source Observability
An Intellyx Brain Candy Brief OpenObserve engineers their observability platform to handle high data volumes with minimal cost. The challenges facing observability tools have never been greater. Gen AI and agentic AI workloads consume and produce tremendous volumes of data at a high rate of speed and create new security risks, incidents, and outages. OpenObserve is engineered to handle high volumes of data and combines multiple solutions in one platform, handling metrics, logs, and traces.
Zymtrace: Optimizing AI Workloads Across CPU and GPU Environments
An Intellyx Brain Candy Brief Zymtrace offers an observability and optimization solution for decreasing AI workload latency and improving GPU/CPU utilization and capacity by conducting deep introspection at the Linux kernel level with runtime eBPF CPU tracing and instrumenting the GPU CUDA kernel.
Opkey Release Advisor: Speed Up Oracle and Workday Release Analysis By 80% with Agentic AI
PRESS RELEASE BY OPKEY “Major enterprise SaaS vendors deliver hundreds of new capabilities every quarter, creating operational complexity for enterprise IT teams who support them,” said Eric Newcomer, Principal Analyst at Intellyx. “Opkey Release Advisor transforms their release documentation into a prioritized adoption plan to help organizations automate the challenge of keeping up with the pace of innovation.” Read the entire press release
Automating Security Incident Response Plans
How much time should you spend creating, reviewing, and maintaining playbooks for responding to security incidents? And how often do you find those playbooks inaccurate or incomplete, no matter what you do? To me, this sounds like a great area for gen AI to provide some assistance. Perhaps the most memorable experience I have with this challenge occurred when I was at Citi and managing a security response team defending against credential stuffing bot attacks.
Show More
loading
Actions
Is this you?
As a journalist, you can create a free Muck Rack account to customize your profile, list your contact preferences, and upload a portfolio of your best work.Get in touch with Eric
Contact Eric, search articles and posts on X, monitor coverage, and track replies from one place.
Learn more about Muck Rack