Yohei Nakajima

Author: Yohei

  • Knowledge Graph 🤝 LLMs

    At Untapped, we pride ourselves on being early in identifying upcoming technology trends, and thought we’d share what we’ve learned recently about the intersection of knowledge graphs and LLMs. For those not familiar, knowledge graphs (Wikipedia) are a type of data representation in the forms of nodes (objects) and edges (relationships). This data structure allows for […]

  • The Future of Autonomous Agents

    I recently did a talk in SF on the future of autonomous agents – then published the deck on X/Twitter here. We’ll be doing an hour long livestream with Q&A on May 16th at 9am on X/Twitter, so mark your calendars (or add your email here for reminders and recording link)! [update: We did the livestream, watch it here.] […]

  • Musings on Semi-Local Inference

    I recently threw out a random thought on Twitter, wondering if there might be room for something I called semi-local inference. This wouldn’t be on-device processing, but something like using a WiFi router to run powerful language models (LLMs). I was curious about the potential benefits—speed, cost, and privacy—over using APIs to power and control […]

  • Impact of BabyAGI

    BabyAGI has been cited in 42 arxiv papers (full list). The following article summarizes these 42 articles. Yohei Nakajima’s project, BabyAGI, appears to have catalyzed a wide range of innovations and research advancements across several domains of artificial intelligence, particularly in the development and application of large language models (LLMs) and agent systems. The impact […]

  • Recent AI Trends and Select Startups 🧠

    First up is AI chips, since Groq made waves this past month with their LLM optimized chips with much faster inference than anything we’ve seen before. Etched is another player in this domain, but we haven’t seen a public demo yet. AI chips aren’t necessary new (eg. Graphcore, Ceerebras), and a question is, will specialized chips maintain value as model architecture evolves? […]

  • PredictiveChat: A Novel Approach to Minimizing Latency in Conversational AI through Anticipation

    *This “paper” was generated by ChatGPT based on the code, author prompt, and X/Twitter discussions. Abstract PredictiveChat introduces an innovative approach to leveraging large language models (LLMs) to anticipate user input, enabling instantaneous response generation in conversational AI systems. By predicting user messages and pre-generating responses, PredictiveChat aims to significantly enhance user experience through reduced […]

  • One agent or multiple agents?

    I recently asked the following question on Twitter: As a consumer would you rather have one personal agent that can book your travel, order you food, help manage your finances, manage your shopping, find you entertainment, keep in touch with friends, track and manage your health, and teach you things OR one specialized for each […]

  • Introducing the GPT VC Associate and the Mean VC 🤖

    Meet our newest team member, the GPT VC Associate. It’s a custom GPT we built that (1) takes pitches from founders (has a specific set of questions it will ask), (2) enriches the info with web browsing, and (3) writes an investment memo which is shared with the founder. While this VC can’t write checks yet, […]

  • Gen AI for Research Market Map

    Research is a domain where AI can bring about a transformative increase in efficiency. Here, we delve into the expansive realm of “General AI for Research,” segmenting it into various markets: Let’s dive into the tools and platforms that are paving the way in each of these sectors: 1. Investment Research 2. Legal Research 3. […]

  • Introducing BabyFoxAGI: The Next Evolution of BabyAGI

    Sorry not sorry, this blog post was AI-generated based on this tweet thread. Ever heard of self-improving task lists, chat UIs with parallel task capabilities, and an assortment of new skills ranging from DALLE to Airtable? Welcome to the world of BabyFoxAGI, the newest mod of the ever-evolving BabyAGI series. This blog post aims to […]