The AiEdge Newsletter
Articles about The AiEdge Newsletter
How Persistent Agents Keep Working
Reconstruct a coding task’s context, preserve accepted work, and recover safely when a worker disappears.
How Self-Improving Agents Learn from Experience
How executable tasks and adaptive curricula turn production failures into useful practice.
How DeepSeek V4.1 Flash Separates Reading and Generation
How a shorter input path, shared attention memory, and selective reconstruction change the cost of long conversations.
Build a Jev-Supervised AI Agent
Typed decisions, guarded tools, and a measured comparison with LLM supervisors
How Jev Turns Questions Into Decisions
Inside the state, typed-output, probability, and application-policy contracts of a decision model
The Complete Guide to Evaluation Pipelines for AI Agents
From production traces and isolated trials to evidence you can use for release decisions
How AI Agent Memory Updates a Changed Preference
Follow one correction from a user message to the active fact a future task retrieves.
How LLM Routers Balance Cache Reuse and Queue Time
Four routing strategies, one repeated prompt, and the point where a cold replica wins
How Test-Time Scaling Allocates LLM Reasoning Across Depth, Width, and Feedback
Why the same inference budget can finish one path, explore alternatives, or learn from a failure
How Constrained Decoding Makes LLM Outputs Follow a Schema
The decoder can guarantee the shape of a tool call while leaving its values completely wrong.
How Sparse Attention Makes Long-Context LLMs Cheaper, and What It Misses
Full reach, sliding windows, and learned selectors turn the same prompt into three different evidence sets.
Four Retrieval Techniques for Video Search
Why frame rate, change detection, temporal hierarchy, and targeted rewatching fail in different ways