AI · Blog — ArceApps

AI · Oct 27, 2026 · 17 min

AgentMemRec: treat agent memory like recommendations

Analyze AgentMemRec's recommender approach to memory retrieval, combining BM25, learned ranking and token-bounded context without extra LLM selection calls.

Read article

#AI

AI · Oct 26, 2026 · 25 min

Graphicle: Code Graph and Memory for AI Coding Agents

Explore how Graphicle combines a code graph, documentation and agent memory; examine its architecture, privacy, benchmarks and closest alternatives.

Read article

#AI

AI · Oct 24, 2026 · 17 min

DeepInfer: what DeepSeek released and what is missing

Explore DeepInfer from DeepSeek: optimized kernels, MoE parallelism, deterministic inference, hardware requirements, and self-hosting limitations.

Read article

#DeepSeek

AI · Oct 22, 2026 · 17 min

Harness²: agents that improve their own harness

Understand Harness² from Google Research: recursive harness optimization, benchmark results, security limitations, and practical lessons for coding agents.

Read article

#AI Agents

AI · Oct 20, 2026 · 17 min

HybridAgents: latent reasoning meets real tools

Explore Hybrid Reasoning Agents: how latent reasoning, continuous working memory, and tools fit together, plus reported gains and practical limitations.

Read article

#AI Agents

AI · Oct 20, 2026 · 16 min

OpenViking: one context layer for multiple AI agents

Explore how OpenViking organizes agent resources, skills and memories with viking:// URIs, L0/L1/L2 retrieval, access controls and shared context.

Read article

#AI

AI · Oct 18, 2026 · 20 min

OpenTelemetry GenAI: make agent behavior observable

Learn how OpenTelemetry GenAI can trace AI agents, tools, memory calls, latency and cost while protecting private data and handling evolving conventions.

Read article

#AI

AI · Oct 10, 2026 · 24 min

DeepSeek Harness Desktop: A Practical Technical Review

Explore DeepSeek Harness Desktop, its plugin architecture, automation, installation, privacy controls and practical trade-offs for indie developers.

Read article

#DeepSeek

AI · Oct 9, 2026 · 24 min

OpenClaw 2.0: A Reliable and Secure Gateway

Learn to run OpenClaw 2.0 reliably with atomic updates, verified backups, Doctor migrations, Tailscale Serve and sensible Gateway security.

Read article

#OpenClaw