The New Stack Podcast28 Mar 2024

LLM Observability: The Breakdown

LLM observability focuses on maximizing the utility of larger language models (LLMs) by monitoring key metrics and signals. Alex Williams, Founder and Publisher for The New Stack, and Janikiram MSV, Principal of Janikiram & Associates and an analyst and writer for The New Stack, discusses the emergence of the LLM stack, which encompasses various components like LLMs, vector databases, embedding models, retrieval systems, read anchor models, and more. The objective of LLM observability is to ensure that users can extract desired outcomes effectively from this complex ecosystem.

Similar to infrastructure observability in DevOps and SRE practices, LLM observability aims to provide insights into the LLM stack's performance. This includes monitoring metrics specific to LLMs, such as GPU/CPU usage, storage, model serving, change agents in applications, hallucinations, span traces, relevance, retrieval models, latency, monitoring, and user feedback. MSV emphasizes the importance of monitoring resource usage, model catalog synchronization with external providers like Hugging Face, vector database availability, and the inference engine's functionality.

He also mentions peer companies in the LLM observability space like Datadog, New Relic, Signoz, Dynatrace, LangChain (LangSmith), Arize.ai (Phoenix), and Truera, hinting at a deeper exploration in a future episode of The New Stack Makers.

Learn more from The New Stack about LLM and observability

Observability in 2024: More OpenTelemetry, Less Confusion

How AI Can Supercharge Observability

Next-Gen Observability: Monitoring and Analytics in Platform Engineering

Join our community of newsletter subscribers to stay on top of the news and at the top of your game.

Denne episoden er hentet fra en åpen RSS-feed og er ikke publisert av Podme. Den kan derfor inneholde annonser.

Episoder(300)

Why CPUs still matter in the age of AI agents

As AI evolves from conversational chatbots to autonomous agents, CPUs are becoming an increasingly important part of the infrastructure equation. In this episode, The New Stack speaks with Bhumik Pate...

11 Aug 26min

Why Doist Says Less AI Can Deliver More

Doist CTO Gonçalo Silva says AI is reshaping software development, but success depends on restraint rather than rapid feature expansion. Instead of chasing every AI capability, Doist prioritizes “subt...

31 Jul 35min

Why your company should (try to) build its own AI SRE

As AI coding agents accelerate software development, they also create new challenges for site reliability engineers (SREs), who are increasingly responsible for debugging systems that no single human ...

30 Jul 30min

Nvidia

In this episode with The New Stack Agents, Frederic Lardinois, NVIDIA’s Joey Conway says advances in AI over the past year have dramatically improved the capabilities of local models, making them prac...

23 Jul 32min

Meet Brain, the AI that decides when Azure is officially down

In this episode, Mark Russinovich, CTO of Microsoft Azure revealed Brain, the AI-powered AIOps system that continuously monitors Azure’s health, detects incidents, identifies root causes, and increasi...

14 Jul 19min

What comes after attention? This startup says it already knows.

Subquadratic is beginning to back up its ambitious claims with benchmarks and third-party validation for its SubQ 1.1 Small model, which uses its proprietary Sparse Attention (SSA) architecture to dra...

7 Jul 20min

“The harness is where the hard work is”: Harness bets on agents that enterprises can trust in production

Harness has introduced Autonomous Worker Agents, a new capability that allows enterprises to replace rigid CI/CD pipeline scripts with AI agents that can deploy applications, run tests, and perform se...

2 Jul 19min

Public cloud vs. on-prem: Summit on where each workload belongs

More than two decades after AWS helped usher in the public cloud era, many organizations are reassessing whether a cloud-first strategy still delivers the cost and operational benefits it once promise...

25 Jun 36min