---
title: "What even is MLOps?"
newsletter: "MLOps Community"
date: 2024-08-22
source: https://aaif.live/newsletters/mlopscommunity/2024-08-22-what-even-is-mlops
---

# What even is MLOps?

*Plus, a panel on getting your day to day quality data, a course on MLOps with Databricks, a looks at agentic AI, LLMs as judges and shrinking LLaMA 3.1 8B.*

*MLOps Community — Agentic AI Foundation, 2024-08-22*

https://x.com/quasa0/status/1823933017217482883

## Design and Development Principles for LLMOps // Andy McMahon // MLOps Podcast #254

It's a bold question, considering the name of the community, but what exactly is MLOps?

The definition I’ve been using comes from Andy, which we discussed along with the evolution of MLOps, highlighting how it has grown to include LLMs and generative AI, especially in regulated industries like banking. He emphasized the need to balance innovation with stability, stressing that leadership and a clear focus on business-driven metrics are essential. He also talked about the importance of infrastructure, monitoring, and treating AI models as commodities within traditional ML workflows.

Now, I just need someone to chat with about defining a boombastic jazz style.

Video [https://home.mlops.community/home/videos/design-and-development-principles-for-llmops] || Spotify [https://open.spotify.com/episode/1nco1R0SoVslseWBYCx0u2?si=C6jNZS9zTAqXvmZta08Pxg] || Apple [https://podcasts.apple.com/us/podcast/design-and-development-principles-for-llmops-andy/id1505372978?i=1000666037976]

[https://podcasts.apple.com/us/podcast/design-and-development-principles-for-llmops-andy/id1505372978?i=1000666037976](https://podcasts.apple.com/us/podcast/design-and-development-principles-for-llmops-andy/id1505372978?i=1000666037976)

[https://home.mlops.community/home/videos/design-and-development-principles-for-llmops](https://home.mlops.community/home/videos/design-and-development-principles-for-llmops)

[https://open.spotify.com/episode/1nco1R0SoVslseWBYCx0u2?si=C6jNZS9zTAqXvmZta08Pxg](https://open.spotify.com/episode/1nco1R0SoVslseWBYCx0u2?si=C6jNZS9zTAqXvmZta08Pxg)

## Data Quality = Quality AI // Panel // AIQCON

## AIQCON Panel

AI without quality data is like coffee without caffeine: ineffective. And both could result in costly mistakes.

This panel focused on the critical role of data quality and the issues poor data quality can lead to, such as model failures and inaccurate dashboards. To avoid these problems, robust monitoring of data metrics like completeness, accuracy, and timeliness were highlighted as essential. The discussion also flagged the importance of end-to-end visibility in data pipelines, the value of data contracts to enforce consistency, and the necessity of breaking down datasets into granular components for better management.

AI without quality data? What’s next, a conference without singing and guitars?

Video [https://home.mlops.community/home/videos/data-quality-quality-ai] || Spotify [https://open.spotify.com/episode/3EVd4JYOdeT887WQ1j4ISd?si=pneMFZDOTRW5_irtCLZ0Rw] || Apple [https://podcasts.apple.com/us/podcast/data-quality-quality-ai-aiqcon-panel/id1505372978?i=1000665625938]

## Job of the Week

[https://www.radai.com/careers?ashby_jid=e33f01ab-6985-4b59-9313-ea5e956b9db0](https://www.radai.com/careers?ashby_jid=e33f01ab-6985-4b59-9313-ea5e956b9db0)

## Hidden Gems

Interested in partnering with us? Get in touch: partners@mlops.community

Thanks for reading. See you in Slack [https://go.mlops.community/slack], YouTube [https://www.youtube.com/channel/UCG6qpjVnBTTT8wLGBygANOQ?view_as=subscriber], and podcast [https://home.mlops.community/public/content/] land. Oh yeah, and we are also on X [https://twitter.com/mlopscommunity]. The MLOps Community newsletter is edited by Jessica Rudd [https://www.linkedin.com/in/jmrudd/].

---
Source: https://aaif.live/newsletters/mlopscommunity/2024-08-22-what-even-is-mlops