sbalnojan

Hey wanted to share a quick few insights we had a MAIA, building yet another AI wrapper (that works as a product).

I’m currently in a product role, but I used to lead marketing—and when our new (and very fresh) marketing lead asked for some structure, I realized I’d never actually written down the process I relied on back then.

So I did. This is the full 6-week focus plan I would’ve wanted in my first B2B SaaS marketing job. It's heavy on frameworks, light on fluff, and painfully honest about where we lost time by doing too much too soon.

Figured it might help others in similar spots.

Data is valuable, but getting it is hard. Here's my take on it, taken from a book I'm writing.

Entrepreneurs often shy away from data-heavy products because they believe they can't make it without the data mountains OpenAI or others have. I don't think that's true. I think, you can:

1. Create value with minimal data, by pulling in data you might not yet own (like external data sets, publically available ones, etc.). Then, use that to attract a minimal set of first users, get their data, and go from there incrementally.

2. You can heavily incentivize early data contributions as Glassdoor does.

3. And you can bootstrap with alternative data sources.

Of course, there are a few battles you shouldn't fight, but I think there's way more potential for data-heavy startups out there than people realize.

My colleague Pat just wrote up an article on LLM-based apps and how to use data engineering tools to help build them faster that blew my mind.

It contains a complete implementation - with scraping context data from a docs website - chunking it, getting embeddings via the openAI API - loading it into pinecone - and finally a simple Q&A interface with streamlit on top of it

Here's a quick summary: - LangChain and LlamaIndex are great tools for quick exploration - But aren't perfect for production-grade use - I think we all know the "LangChain is pointless" debate, but there's a lot of real meat to it, and Pat describes a few of them (a lot of LangChains extractors are super basic, 2-3 liners without retries etc.) - LLM applications are all about moving data, extracting and enriching data (creating embeddings!) are the most expensive ones of those steps - A bunch of data engineering tools are out there that make these two steps much easier, versionable, robust, and reproducible. - Meltano is one such tool and Pat implemented the above described pipeline with it

FWIW: The GitHub project that comes with the post is super easy to run and super modular. I just tested it and was able to modify everything for my own application within 30 mins.