
Baptiste Jamin
@baptistejamin
Followers
2K
Following
2K
Media
351
Statuses
4K
Co-Founder/CEO @crisp_im - Speaking about AI, LLMs & code. - Bootstrapping & coding since 2015 the future of customer service.
Nantes, France
Joined June 2010
Over the past few weeks, I fine-tuned different models for @crisp_im to enhance the RAG pipeline and make it on steroids.
1
0
1
RT @attacomsian: Set up @crisp_im live chat for Depost AI (both marketing + dashboard sites). It's a great way to collect feedback and ta….
0
3
0
Over the past two weeks, we released at @crisp_im two new ML models, trained state-of-the-art foundation models. For our support, we are increasing the percentage of auto-answerable questions from 33% to 60%.
0
1
12
RT @bdkjones: Apple Music in macOS 26. Primary controls go on the bottom for iOS apps because that's where your thumbs are. On Mac, contro….
0
1
0
RT @crisp_im: Part of the team from across the globe has started to join us in Nantes to celebrate our 10 years anniversary .
0
3
0
RT @marc_louvion: I just set up @crisp_im for my startup and I get 7x more requests than a typical email "support" button. This is gold if….
0
6
0
We just released a new layout for Inboxes on @crisp_im , made by the great and unique @eliottvct . It allows to manage Crisp Inbox in a different way, for users being more familiar with email solutions.
0
1
4
Qwen 3 , 4B variant overtakes GPT4o-mini on the Arena Hard benchmarks. It seems now a model that can fully replace Llama 3 8b.
Introducing Qwen3! . We release and open-weight Qwen3, our latest large language models, including 2 MoE models and 6 dense models, ranging from 0.6B to 235B. Our flagship model, Qwen3-235B-A22B, achieves competitive results in benchmark evaluations of coding, math, general
0
0
2