# EP 142 - AI Tools: LLMs with Sani Manic

Page: https://stenobird.com/podcast/product-for-product-management-3487998/ep-142-ai-tools-llms-with-sani-manic
Text version: https://stenobird.com/podcast/product-for-product-management-3487998/ep-142-ai-tools-llms-with-sani-manic.md
Podcast: [Product for Product Management](https://stenobird.com/podcast/product-for-product-management-3487998)
Published: 2025-11-26T06:00:09+00:00
Episode link: https://www.spreaker.com/episode/ep-142-ai-tools-llms-with-sani-manic--68525293
Audio file: https://api.spreaker.com/download/episode/68525293/ep_142_sani.mp3
Processing state: not_requested
JSON: https://stenobird.com/v1/public/podcasts/product-for-product-management-3487998/episodes/ep-142-ai-tools-llms-with-sani-manic
Duration seconds: 2733

## Resource

We continue our AI Tools series with a deep dive into using Large Language Models (LLMs) for research, featuring Slobodan (Sani) Manić, AI skeptic, podcaster, and founder of the AI Fluency Club. Sani joins Matt and Moshe to share why context, careful prompting, and critical thinking are essential for getting real value out of today’s LLMs in product work. Drawing on his work as a product builder, educator, and host of No Hacks Podcast, Sani challenges common myths about AI’s capabilities and underscores both its practical uses and its risks for product managers. The conversation ranges from practical workflows to future visions of invisible AI, open-source models, and the real state of the “wrapper economy” built on major LLM providers. Join Matt, Moshe, and Sani as they explore: - Why most LLM workflows boil down to two mindsets: understanding your work, or avoiding understanding it - The crucial role of context and authority, why careless prompting leads to hallucinations, and how to break questions into smaller steps for better results - How LLMs fit as accelerators for deep research, surfacing insights faster than classic search engines, but always requiring fact-checking - Why Sani uses Google’s Gemini and NotebookLM, and the value of integration with your company’s existing tools - The open-source LLM alternative: privacy, flexibility, and why some see this as the future for secure enterprise AI - Pitfalls of the “wrapper economy,” vendor lock-in, and shaky business models based on reselling tokens - Starting out: how to include LLMs in PM research without reinventing your workflow, and why you must be careful with company data - The risks and limitations of AI today, especially in enterprise and sensitive environments - How internal AI context in tools like Atla…

## Actions

- request_transcript: `POST https://stenobird.com/v1/public/podcasts/product-for-product-management-3487998/episodes/ep-142-ai-tools-llms-with-sani-manic/transcription-requests` — Idempotently request low-priority transcript generation for this episode.
- read_markdown: `GET https://stenobird.com/podcast/product-for-product-management-3487998/ep-142-ai-tools-llms-with-sani-manic.md` — Read the agent-friendly Markdown representation of this episode resource.

A page view does not enqueue transcription. Agents should invoke `request_transcript` explicitly when they need this episode processed.

## Transcript

Full transcripts are not published on public pages unless there is a clear rights basis.