cd /news/ai-agents/hey-im-sharing-an-early-stage-experi… · home topics ai-agents article
[ARTICLE · art-133978] src=discuss.huggingface.co ↗ pub= topic=ai-agents verified=true sentiment=· neutral

Hey! I’m sharing an early-stage experimental project: SelMem

Developer Jibs79 released SelMem, an early-stage experimental Rust project exploring whether LLM agents can develop persistent behavioural divergence through a reconstructive memory system that forgets, distorts, and consolidates experiences over time. The project claims two agents with identical histories can end up with different internal narratives and personalities, though Jibs79 states there are no serious benchmarks yet and is seeking feedback on evaluation and comparison to existing memory architectures.

read1 min views1 publishedSep 18, 2026
Hey! I’m sharing an early-stage experimental project: SelMem
Image: Discuss (auto-discovered)

Jibs79 1

It explores whether LLM agents can develop persistent behavioural divergence through a reconstructive memory system that forgets, distorts, and consolidates experiences over time. Two agents with identical histories can end up with different internal narratives and “personalities.”

Written in Rust. Very early but promising, no serious benchmarks yet.

Repo: GitHub - jbsalles/Selmem: Selective memory for LLMs · GitHub Feedback welcome, especially on evaluation and comparison to existing memory architectures.

── more in #ai-agents 4 stories · sorted by recency
── more on @jibs79 3 stories trending now
sponsored brought to you by zahid.host 4,200+ EU-deployed projects
reading about agents? ship yours in a single git push.

Run your AI side-project on zahid.host

EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.

$git push zahid main
Live at https://your-agent.zahid.host
Get free account → Pricing
from €0/mo · no card required
LIVE [news/hey-im-sharing-an-ea…] indexed:0 read:1min 2026-09-18 ·