It’s been 4 months since the start of the autonomous agent revolution, let’s check some out. We’ll start with BabyAGI and AutoGPT, two of the inaugural autonomous agents. AutoGPT has a community of devs building lots of integrations, while BabyAGI has a handful of mods that Yohei has released. Other general agents include Adept who’s raised the most, beebot who has scored highly on benchmarks, and Godmode, with heavy user-in-the-loop. I was also excited to see LayerBrain, for moving data across platforms.
Cognosys and Aomni are two leading research agents I’ve covered, and there’s also GPT-Researcher. Recently trending is MULTI-ON for it’s ability to handle browsers (like booking a flight), which Hyperwrite’s assistant can do as well. We’re seeing more personal assistants for scheduling and email like Lindy.ai and Floode, and for keeping CRMs up to date like Linc. Tools for deployment of agents include Fine-tune, Superagent, and e2b.dev – they have a one-click deployment of smol-ai, a popular coding agent. Excited for all the specific agents like AutoRPG for generating game levels, Diagram for design, Air for customer service/sales, and The Simulation for AI generated TV shows and more.
Tons of approaches to coding agents that involve various approaches to multi-step thinking of architecture, features, environment, to generate entire code bases based on a prompt, with various opinionated approaches on what info to collect ahead of time, where to get user feedback, and more. Some ones that trended recently include smol-ai, GPT-Engineer, and Clippy (AI coding teams). You can see a larger list of AI agent projects here.