longmemeval experiment arch: 1) deterministic ingestion/extraction (85.6% accuracy, 86.2% retrieval)
2) semantic ingestion/extraction (84.8% accuracy, 94.9% retention)
3) semantic ingestion/deterministic extraction (87.6% accuracy, 90.0% retrieval) this was staggered, not
@yoheinakajima
-
Longmemeval Experiment: Three Ingestion/Extraction Configurations Compared
By
–
-
Claude kept telling me to go to bed during evals
By
–
was running some evals this weekend and claude kept trying to get me to go to bed
-
Experimenting with Event Sourced Agent Runtime
By
–
experimenting with a new event sourced agent runtime that i made 🙂
-
Cofounder as AI cofounder for ActiveGraph, managing website and blog
By
–
for ActiveGraph, i'm using Cofounder by @intelligenceco as my… cofounder I'm working on the repo, doing research. Cofounder built the website, manages newsletter/blog, and when I do a new blog or have something new to share, I just ping it saying "here's a link/content, can
-
Active Graph video: flipped agent architecture, blackboard system, self-improvement
By
–
longer form video on Active Graph [7 min 22 sec]
— Yohei (@yoheinakajima) 21 mai 2026
– flip the agent architecture
– 1970s blackboard system
– rollback, fork, diff agent runs
– experience + behaviors + beliefs = you
– behaviors can write behaviors (self-improvement)
github: https://t.co/8ENyNt4gYD pic.twitter.com/WeN2D60bdjlonger form video on Active Graph [7 min 22 sec] – flip the agent architecture
– 1970s blackboard system
– rollback, fork, diff agent runs
– experience + behaviors + beliefs = you
– behaviors can write behaviors (self-improvement) github: https://
github.com/yoheinakajima/
activegraph
… -
Active Graphs: A New Approach to Agent Architecture
By
–
we've been building agents around the llm, starting with conversations, adding tools, giving rules, logging everything, and storing a form of it as retrievable state
— Yohei (@yoheinakajima) 20 mai 2026
Active Graphs flips this. what if we treated the logs as the agent, which include rules provided (and changed),… https://t.co/RLpP6SOynQwe've been building agents around the llm, starting with conversations, adding tools, giving rules, logging everything, and storing a form of it as retrievable state Active Graphs flips this. what if we treated the logs as the agent, which include rules provided (and changed),
-
AI-Powered Development Workflow: From Ideation to Testing
By
–
for anyone curious, this was the result of many experiments bouncing around but this version uses… initial ideation: chatgpt for ideas/
@replit for quick build
final repo build: opus 4.7 prompting claude code
pypi testing: ran so many tests w @replit who wrote reports that I -
Development progress on LLM tool calling integration
By
–
thank you, yeah, there's still a lot i need to do around the LLMs. getting the underlying stuff working was most of the work up to now. thx for the pointer on model, will update when i fix tool calling
-
Technical Analysis Request for LLM-based Agent Architecture
By
–
@grok @perplexity_ai review http://
docs.activegraph.ai, analyze the approach critically, and explain in detail where this sits in the evolving landscape of LLM based agent architecture