Lookio
Build knowledge-based assistants. Call them via API.
70 followers
Build knowledge-based assistants. Call them via API.
70 followers
Lookio lets you build AI assistants that answer questions using your company's documents. Perfect for automating customer support, scaling content creation, and powering workflows. API-first design with flexible query modes. Start free with 250 credits.
Free
Launch Team / Built With
Render: Get up to $100K in startup creditsCloud infrastructure for founders who want to move fast and NOT break things
Promoted
Hey Product Hunt! 👋 I'm excited to share Lookio, the platform that makes reliable and sourced AI knowledge retrieval (RAG) accessible for every company.
I spent the last 2 years testing all possible ways to nail content workflows based on the company's documentation (previously in the carbon accounting industry), especially for marketing use case when the team struggles to write expert content at scale.
Building accurate and reliable RAG systems is an art (cleaning & formatting the data, embedding pipeline, metadata, chunking methods, retrieval techniques...) and I wanted to offer an easy-to-implement solution.
Early users describe it that way "NotebookLM, but with an API".
3 steps to get started:
- Upload your resource documents (PDF, Docx, URLs, plain text...)
- Configure Assistants (context, output guidelines, resources access)
- Query your Assistants via API (e.g. via n8n, Make, Zapier) to automate retrieval
Main use cases are sourced SEO content generation, support chatbots, internal tools, but high quality knowledge retrieval can be embedded into many more company workflows.
I also made sure to make the tool accessible so you don't need a subscription and can simply pay as you go with credit packs!
We're offering 250 free credits to get started – no credit card needed!
Shipping improvements on a weekly basis - we'd love to hear your feedback and see how you're building with Lookio.
Big shoutout to Softr, no-code app builder, for powering the tech behind the platform that allows us to evolve so fast
Thanks for checking us out!
Nice one! does it work with website pages and documents only ? Can we ingest other sources like intercom messages history?
@arturmkrtchyan good question - I've released an API endpoint to import resources and it supports files (PDF, docx), URLs to fetch, but also any plain text which can be Intercom messages that you programmatically send via API.
Hope this helps 🙏
Great Tool from a great builder.
Is it possible to connect Lookio with airtable ?
@julienairtable great question - and the answer is YES! You'll have to use the script module in Airtable automations to run the API call and the output contains a key with the result that you can then map to create or update a record in your base :)
I remember when you had your first version (I guess only weeks ago!?) that already at this stage scored 37/40 possible points in the Agentic Arena Community RAG Challenge of @n8n 👏 can't wait to give it another spin!
@geckse exactly, that n8n challenge was a great benchmark and I could actually work on it to go from 33 to 37 points with further optimizations 🔥
This looks super useful, especially for automating content workflows. How well does Lookio handle mixed data sources like PDFs + URLs in the same assistant context?
Congrats on the launch@guillaume_duvernay! Excited to hunt this product —first, for its being amazing and so valuable for many companies, and second, for being built on Softr!
I hope more companies leverage this tech and wish you good luck with this launch! 🔥
@mariam_hakobyan5 when building this API-first product, what I needed was an interface that was the most intuitive possible for users to setup their workspace, import their resources, configure their assistants, get their API key and then track their usage.
And Softr exactly allowed that - super reliable blocks and modern interface - also perfect for EU-based data!!
There's RAG then RAG - it's all in the details. Looking forward to testing this out for scenarios where a custom pipeline doesn't make sense.
@max_tkacz 100% aligned - not every use case requires it's own fully custom RAG data pipeline, sometimes you're happy to enjoy excellent retrieval with one single node, so you can actually focus on integrating that well into your business workflows and flowgram smoothly