I’ve heard about Elicit from the Latent Space podcast back in April where Jungwon and Andreas told their founding stories. I was surprised to hear that they had researched trying to emulate advanced AI with humans and learned a lot about topics around segmented intelligence, which I later experienced while building LLM applications. I also found another episode insightful where James explained the principles of AI engineering is a fault-tolerant mindset and distributed system skills.
I gave a workshop on Flow Engineering(designing and optimizing LLM workflows) at AI Engineer Summit 2025 in NYC.
I’m interested in unbundling artificial intelligence into real world tasks. LLMs are already so good to understand almost any unstructured data given a proper context information. However, it’s too ephemeral; it’s a frozen, stateless, unreflective machine that only wakes up for a sliver of time.
In order to utilize the full potential of this new intelligence, we need to build a scaffolding around it so that it can complete complex tasks by breaking down into sub tasks and creating workflows. Moreover, the intelligence should serve humans’ request, meaning we need to invent a new UI UX and systems for alignments.
I admire the founders of Elicit for their work in this regard, and I want to contribute to the future journeys at Elicit.
But I'll be a little honest here: I like the people at Elicit, and that's what mostly made me interested in this role. Although I don't know all of you personally, I've watched several interviews and felt that the founders and engineers at Elicit are earnest and ethically driven, which I care about most when joining a team.