Christophe Pasquier

Powers 90% of Super intelligence.

Incredible tool, we used it to generate our Super 2.0 launch video, :kiss_chef:

I must have received 20+ messages from friends askign me how I make my demo videos by then. Screenstudio is one of the most delightful and well thought product I've used in the last 5 years. An absolute killer product.

Langgraph helped build Super by providing a standardized framework that made the complex LLM/source interactions more manageable through a declarative graph structure. It enabled better parallelization which significantly improved response speed, while also providing full graph state/sequence debugging capabilities that made troubleshooting easier. The migration to Langgraph unlocked new capabilities like human-in-the-loop pauses for clarification and better fallback mechanisms when models fail. Additionally, Langgraph's functional API allowed developers to write more natural, sequential code for complex subgraphs while still maintaining the benefits of the graph framework.

Claude by Anthropic has been instrumental in building Super. The team chose Claude as the primary language model because of its superior reasoning capabilities and contextual understanding. Claude powers Super's core features including cross-platform search, source explanations, and document analysis with remarkable accuracy. The development team made specific prompt engineering improvements for Claude, including better system prompting and JSON handling to maximize its performance. Claude's ability to handle complex queries and provide chain-of-thought reasoning has enabled Super to deliver 8x more relevant results than regular AI search. We considered OpenAI or Gemini, but we use them as fallbacks.

Elasticsearch serves as the backbone of Super's search capabilities, enabling fast and accurate retrieval of information across diverse data sources. Its flexibility allows us to handle complex queries generated by our AI system, while providing robust analytics on usage patterns. The graph-based architecture we built integrates Elasticsearch queries with LLM prompts, creating a powerful system that can search, analyze, and synthesize information. Elasticsearch's scalability and customization options have been crucial for local development, error handling, and continuous improvement of Super's performance. Simply put, Elasticsearch's speed and adaptability made it the ideal foundation for building Super's advanced knowledge retrieval system.

Absolutely massive, really great job on the past months @mboudoukhane and team. The product got soo freaking clean and polished, but more importantly the value prop is so much clearer than in the past, when you land on the app, the dots connect in your mind instantly. Really amazing to see, great job!!

So nice to see proper OSS libraries challenging Stripe Billing. It's an essential part of any online business, having open source alternatives feels like a basic need for our industry. GG Lago team!!