IT Ops Query

AWS VP plans Amazon Bedrock Agents observability and more

The leader of search, observability, and cloud operations at Amazon Web Services outlines her strategic perspective on monitoring AI agent systems such as Amazon Bedrock Agents. Ultimately, she envisions unified visibility, including security, across AI workflows, from silicon to application, both inside and outside AWS.
Featuring: Nandini Ramani, Vice President at Amazon Web Services
In today’s episode, we’ll cover…
Generative AI for telemetry analysis
Distributed tracing for AI agents
Observability and the future of DevSecOps
and more!
References:
Amazon Q, Bedrock updates make case for cloud in agentic AI
Amazon Bedrock users adapt app dev to GenAI
More reasoning, interoperability key to future of agentic AI
AWS Enhances Cloud Security With Better Visibility Features
To learn more about IT Operations, check out SearchITOperations.com.
To watch clips from our podcast, subscribe to our YouTube channel, @EyeOnTech.

Dancing with AI: AI risks, rewards and the evolution of SREs

Experienced IT practitioners offer their predictions for how AI will change observability, the role of the site reliability engineer and incident management within enterprise organizations, from best-case scenarios to pitfalls and risks. Come for the 40 combined years of IT ops expertise, stay for the "Highlander" references and football analogies.
Featuring: Jonathan Moore, domain architect, Progressive Insurance and Laura Vetter, chief technologist and co-founder, Evolutio
In today’s episode, we’ll cover…
AI risks
AI agents
Site Reliability Engineering
and more!
References:
Datadog AI agent observability, security seek to boost trust | TechTarget
SREs map uncharted territory with LLMOps | TechTarget
ServiceNow shops share AI copilot results, prep for agents | TechTarget
More on observability from Jonathan Moore's Splunk .conf 2024 presentation
To learn more about IT Operations, check out SearchITOperations.com.
To watch clips from our podcast, subscribe to our YouTube channel, @EyeOnTech.

Datadog AI researcher teaches time-series models new tricks

Ameet Talwalkar joined observability vendor Datadog in February to lead a new AI research lab at the company. He is also an Associate Professor at Carnegie Mellon University. One of the first projects from the lab was released to open source in May – Toto, a foundation model trained on real-world observability data. Talwalkar discusses how Toto could improve observability forecasting tools such as Datadog's Watchdog, and does some forecasting of his own about how AI agents will alter observability and human-computer interactions.

The good, the bad and the ugly of AI and observability

Steve Koelpin is a seasoned engineer and award-winning data strategist who specializes in observability, logging, and data pipelines at high scale. In his experience, generative AI in observability tools can help reduce the time it takes to resolve IT incidents, but can also foment dependence on a 'black box' tool. There's also the matter of feeding AI the large amounts of data it requires without breaking the bank, where Koelpin shares his dos and don'ts.

Enterprises adapt to generative AI operations

Nancy Gohring is a senior research director at IDC, focused on big picture trends related to enterprise AI adoption, including business, organizational and technology architecture transformation in the context of AI and GenAI. In this wide-ranging interview, Gohring gives the big-picture view of the challenges and changes GenAI represents for enterprise IT operations teams, from roles to teams to tools.  

GenAI and business transformation

Enterprise Strategy Group analyst Torsten Volk says he spends so much time playing with generative AI tools that "if I used fewer, I'd get a lot more done." But judiciously used -- and with plenty of human oversight -- LLMs and AI agents can be catalysts for fundamentally changing businesses, from application modernization to massive data processing, including the growing volume of telemetry gathered by observability tools. 

Get ready to meet your new digital coworkers

One of the enterprise tech industry's most influential voices on all things AI, independent analyst Andy Thurai, says in a wide-ranging discussion that IT pros should prepare to work with AI agents in the very near future, despite unanswered questions about their orchestration and the reliability of their decision-making. 

Mitigating AI's unique risks

"AI is not just another technology layer; it's a distinct stack with its own complexities and risks," according to Coralogix CEO Ariel Assaraf. In this episode, Assaraf details those unique issues and how his company plans to address them using its recent acquisition, Aporia.

StarTree founder positions Apache Pinot for AI observability

Kishore Gopalakrishna says that the observability market is ripe for disruption in the AI age, particularly at the storage and query layer. The former LinkedIn architect is now co-founder and CEO of StarTree, which offers a real time analytics platform based on Apache Pinot. The company has a preview version of its StarTree Cloud service for observability that supports metrics, logs and traces, and shipped StarTree ThirdEye last year for anomaly detection and root-cause analysis. In this episode, Gopalakrishna discusses why the new products are a cheaper, faster alternative for observability.

LLMs vs AIOps: have we seen this movie before?

Gartner analyst Gregg Siegfried reflects on the previous generation of AIOps tools that never quite lived up to their "NoOps" promise, looks at how generative AI tools have already affected IT ops pros' day to day work, and predicts the ways AI agents are poised to even more dramatically alter the observability and IT automation landscape.