honcho

https://github.com/plastic-labs/honcho has the idea of one sided observations for RAG.

Hey HN, It’s Vineeth from Plastic Labs. We've been building Honcho, an open-source memory library for stateful AI agents. Most memory systems are just vector search—store facts, retrieve facts, stuff into context. We took a different approach: memory as reasoning. (We talk about this a lot on our blog) We built Neuromancer, a model trained specifically for AI-native memory. Instead of naive fact extraction, Neuromancer does formal logical reasoning over conversations to build representations that evolve over time. Its both cheap ( $2/M tokens ingestion, unlimited retrieval), token efficient and SOTA: LongMem (90.4%), LoCoMo (89.9%), and BEAM. On BEAM 10M—which exceeds every model's context window—we hit 0.409 vs prior SOTA of 0.266, using 0.5% of context per query. Github: https://github.com/plastic-labs/honcho Evals: https://evals.honcho

Plastic Labs | DevRel, Summer Interns| NYC | https://plasticlabs.ai Plastic Labs is building systems that map personal identity using AI-native memory & social cognition. This means individually-aligned agents that you can trust to act autonomously on your behalf & agents with rich identities all their own. We're pursuing this with our flagship product Honcho (github.com/plastic-labs/honcho) a memory & identity layer for agentic applications. learn more at https://blog.plasticlabs.ai We are looking for a Developer Relations Engineer to help us evangelize Honcho to the broader dev community. This involves engaging the community online & in-person, building demos, writing documentation & blog posts, and working with the dev team to communicate feedback from the community. Check out the full requisition here: https://blog.plasticlabs.ai/car

Plastic Labs | NYC | Various Engineering | Full-time | https://plasticlabs.ai Plastic Labs is a research driven product company working to radically decentralize alignment for AI. Our flagship product, Honcho (https://honcho.dev), serves as a bridge between users and AI systems. Honcho creates detailed user profiles that instantly customize AI experiences across multiple applications, ensuring each person's AI tools work according to their specific preferences and needs. - Tutor-GPT: https://github.com/plastic-labs/tutor-gpt - YouSim: https://yousim.ai - Honcho: https://github.com/plastic-labs/honcho Learn more on our blog: https://blog.plasticlabs.ai/careers/Work-at-Plastic We are hiring a variety of full time roles: - Full-Stack Engineer: https://blog.plasticlabs.ai/careers/Full-Stack-Engineer - Platform Engineer: https://blog.plasticla

Plastic Labs | NYC or Remote (Americas) | Full Time | 100-200k + equity Plastic Labs is a VC backed research-driven company solving identity for the agentic world. We’re building Honcho, the social cognition layer for AI-powered applications. Honcho synthesizes high-fidelity user representations to instantly personalize the UX of entire app ecosystems—individually aligning each user’s agent stack. Some of the things we've built - Tutor-GPT: https://github.com/plastic-labs/tutor-gpt - YouSim: https://yousim.ai - Honcho: https://github.com/plastic-labs/honcho Learn more on our blog: https://blog.plasticlabs.ai/careers/Work-at-Plastic We are hiring a variety of full time roles: - Full-Stack Engineer: https://blog.plasticlabs.ai/careers/Full-Stack-Engineer - Founding Engineer: https://blog.plasticlabs.ai/careers/Founding-Engineer - ML Researc

Working on Honcho[0] a platform for personalizing LLM based applications to individual end users. It was inspired when working on a tutoring service and realizing that the key problem of AI tutors was that they didn't know the student well enough. As we started tackling the problem other builders around us mentioned similar concerns. It's been pretty hard to explain the value to those not experiencing the same problems, but it's slowly catching on. [0] https://github.com/plastic-labs/honcho