Similar to This
Previous

Paid articlePaywall Google adds two new specialized chips for AI agents and security to 8th-gen TPU lineup — TPU 8i Max puts two chips on one package for long-context work ...
tomshardware.com/tech-industry/artificial-intelligence/google-adds-two-new-specialized-chips-for-ai-agents-and-security-to-8th-gen-tpu-lineup-tpu-8i-max-puts-two-chips-on-one-package-for-long-context-work-tpu-8tx-swaps-arm-hosts-for-amd-e

Why AWS wants developers to stop babysitting their AI agents: AWS’s new Kiro bots keep their own memory and run parallel agent sessions. The promise: developers ...
thenewstack.io/kiro-bots-agent-memory/

Cohere offers to put agents in lockdown mode with strict ACLs: Like OpenClaw but with access control
theregister.com/ai-and-ml/2026/10/05/cohere-offers-to-put-agents-in-lockdown-mode-with-strict-acls/5301219

AWS open-sources an AI agent it says is 45% cheaper than Claude Code and Codex: Strands Harness bundles the tools, context and memory developers otherwise build ...
thenewstack.io/aws-strands-harness-agent/

AWS's OpenSearch pitch ahead of OpenSearchCon | VentureBeat: What is AWS pitching for OpenSearch? Retrieval, serverless costs and agent memory — plus what the ...
venturebeat.com/data/awss-opensearch-pitch-ahead-of-opensearchcon-retrieval-serverless-costs-and-agent-memory

Your agent context needs a development lifecycle: Manage your AI agent context like software. Discover how the Context Development Lifecycle improves skill ...
thenewstack.io/agent-context-development-lifecycle/

MinIO pitches persistent memory for agents with work to finish: AIStor keeps context, files, and secrets under customer control so interrupted jobs can pick up ...
theregister.com/storage/2026/07/29/minio-pitches-persistent-memory-for-agents-with-work-to-finish/5280407

Couchbase brings persistent agent memory to the edge | VentureBeat: Couchbase AI Data Plane combines persistent agent memory, vector search and an enterprise MCP ...
venturebeat.com/data/ai-agents-need-context-everywhere-they-run-even-where-the-cloud-cant-follow

Your enterprise AI agents should automatically remember which model is right for which task. Mindstone built the capability with Rebel.
venturebeat.com/orchestration/your-enterprise-ai-agents-should-automatically-remember-which-model-is-right-for-which-task-mindstone-built-the-capability-with-rebel

Hypernetworks and AI agent autonomy explained | VentureBeat: Why AI agents stall in production: fine-tuning forgets, RAG leaks context. Hypernetworks generate a ...
venturebeat.com/orchestration/fine-tuning-forgets-rag-leaks-context-hypernetworks-build-the-model-your-agent-needs-on-demand

LLM agent memory at 0.12% of model parameters: Researchers built delta-mem to give AI agents working memory at 0.12% parameter overhead ...
venturebeat.com/orchestration/a-0-12-parameter-add-on-gives-ai-agents-the-working-memory-rag-cant

Enterprise AI agents fail because they forget: RAG retrieves documents but not decision logic, causing agents to act on expired rules. Decision context graphs ...
venturebeat.com/orchestration/enterprise-ai-agents-keep-failing-because-they-forget-what-they-learned

GitHub - kevdogg102396-afk/packrat: Auto-learning codebook compression for AI agent memory. Makes context files 2-10x smaller while staying fully readable by ...
github.com/kevdogg102396-afk/packrat

GitHub - Rtalabs-ai/aura-research: LLM-powered research knowledge base — compile raw documents into a living wiki with persistent agent memory and RAG ...
github.com/Rtalabs-ai/aura-research

GitHub - ekailabs/contexto: Context Engine for your AI agents.: Context Engine for your AI agents. Contribute to ekailabs/contexto development by creating an ...
github.com/ekailabs/contexto

GitHub - kevin-hs-sohn/memaware: Benchmark for measuring memory awareness in AI agents — the ability to surface relevant past context without being asked
github.com/kevin-hs-sohn/memaware

GitHub - DigitalMeatbag/lambertians: Locally-running AI agent with lifecycle mechanics: bounded lifetimes, episodic memory, pain signals, and mortality.
github.com/DigitalMeatbag/lambertians

GitHub - srimallya/subgrapher: Subgrapher is a local-first semantic browser workspace where references become runnable memory apps with graph context ...
github.com/srimallya/subgrapher

GitHub - ptobey/local-memory-mcp: Local-first personal RAG memory system for AI assistants via MCP. Stores text-first chunks with lightweight metadata ...
github.com/ptobey/local-memory-mcp
Next
