In this episode of Building With AI, host Haroon Choudery is joined by Jason Liu, a seasoned AI operator and consultant who advises seed-stage start-ups on tech strategy, research, and infrastructure.
With several years of experience in machine learning, Jason has gained recognition for creating AI-specific dev tooling and contributing to several open-source projects.
During the conversation, they discuss:
- General applications vs. AI applications,
- How developers should work with LLMs,
- What LLMOps companies need to focus on,
- Why saying 'no' can improve your AI's success rate,
- Building agile AI systems,
- Best practices for RAG systems,
- And much more!
—
[00:00] Up next: Jason Liu
[00:25] Episode Introduction
[01:12] Autoblocks: AI Optimization Platform
[02:07] Jason's introduction
[03:44] General applications vs. AI applications
[05:08] The two categories of dev tools
[06:43] The AI landscape for non-technical users
[08:47] The one skill developers need
[09:57] The eval puzzle
[11:52] What LLMOps companies need to focus on
[16:26] Why saying 'no' can improve your AI's success rate
[18:15] How can you build an agile AI system?
[21:41] Strategies for building products that stick
[24:29] Similarities between RAG and recommendation frameworks
[26:03] Best practices for RAG systems
[28:27] The importance of short-term metrics
[29:30] Data strategies for effective RAG systems
[35:22] Jason's hottest take on AI
[36:22] Jason's future projects
—
Brought to you by Autoblocks (https://www.autoblocks.ai) — how product teams deliver better AI experiences.
—
Subscribe to Building With AI:
Where to find Jason Liu:
Where to find Haroon Choudery:
Learn more about Autoblocks: