We Can All Be AI Engineers and We Can Do It with Open Source Models // Luke Marsden // #273

Key takeaways

Anyone can become an AI engineer by leveraging open-source models and embracing modern tools.
CI/CD principles apply to AI applications, highlighting the need for thorough testing and deployment cycles.
Establishing an AI specification (AISpec) is essential for managing generative AI applications effectively.
Non-technical users can prototype AI applications using user-friendly web interfaces, democratizing AI development.
Testing and constant evaluation of AI integrations, like JIRA's, is vital to ensure functionality and reliability.

Summary

Introduction to AI Engineering

In this episode, we dive into the transformative realm of AI engineering with Luke Marsden, the CEO of HelixML. Luke stresses that the journey to being an AI engineer doesn't require a Ph.D. but rather an open-source mindset and familiarity with modern tools. He reflects on his journey with the MLOps Community, which arose from the need for tracking and lineage in AI applications.

Understanding CI/CD for Generative AI

Luke dials into the crucial components of continuous integration and continuous delivery (CI/CD) when it comes to generative AI applications. He emphasizes that AI models should be treated similarly to software, needing rigorous testing and deployment methodologies. He introduces the idea of utilizing eVALS to test generative applications effectively, drawing parallels to traditional software systems.

The AI Specification (AISpec) Project

One of the cornerstones discussed is the AISpec project, which aims to create standards for AI application specifications, much like how traditional applications use YAML manifests. This standardization allows developers to manage AI apps as they would conventional software. Luke explains how these specs can incorporate everything from model definitions to API integrations.

Empowering Non-Technical Users

An interesting aspect of the conversation is Luke's vision for non-technical users to create their own prototypes using open-source models. He envisions a web interface akin to GPT's feature allowing individuals unfamiliar with deep technical jargon to build AI applications effectively. This bridges the gap between those who can code and those who conceptualize the applications.

Real-World Implementation & Learnings

Luke shares practical experiences from building a JIRA integration, showcasing how to construct prompts, manage API calls, and summarize responses. He discusses the importance of testing, particularly through the lens of the AI spec to ensure that each new feature or change does not regress previously verified functionalities.

Show notes

Luke Marsden, is a passionate technology leader. Experienced in consultant, CEO, CTO, tech lead, product, sales, and engineering roles. Proven ability to conceive and execute a product vision from strategy to implementation, while iterating on product-market fit.

We Can All Be AI Engineers and We Can Do It with Open Source Models // MLOps Podcast #273 with Luke Marsden, CEO of HelixML.

// Abstract

In this podcast episode, Luke Marsden explores practical approaches to building Generative AI applications using open-source models and modern tools. Through real-world examples, Luke breaks down the key components of GenAI development, from model selection to knowledge and API integrations, while highlighting the data privacy advantages of open-source solutions.

// Bio

Hacker & entrepreneur. Founder at helix.ml. Career spanning DevOps, MLOps, and now LLMOps. Working on bringing business value to local, open-source LLMs.

// MLOps Swag/Merch

https://mlops-community.myshopify.com/

// Related LinksWebsite: https://helix.ml

About open source AI: https://blog.helix.ml/p/the-open-source-ai-revolution

Ratatat Cream on Chrome: https://open.spotify.com/track/3s25iX3minD5jORW4KpANZ?si=719b715154f64a5f

--------------- ✌️Connect With Us ✌️ -------------

Join our Slack community: https://go.mlops.community/slack

Follow us on Twitter: @mlopscommunity

Sign up for the next meetup: https://go.mlops.community/register

Catch all episodes, blogs, newsletters, and more: https://mlops.community/

Connect with Demetrios on LinkedIn: https://www.linkedin.com/in/dpbrinkm/

Connect with Luke on LinkedIn: https://www.linkedin.com/in/luke-marsden-71b3789/