The journal

Notes from the context layer.

Engineering deep-dives, benchmarks and customer stories on persistent code context, knowledge graphs, and the AI coding tools we work with daily.

All 86 Customers 20 Engineering 66

21–30 of 86 · page 3 of 9

May 12, 2026 LLM benchmarks code analysis DeepSeek V4

We Benchmarked 14 LLMs for Code Analysis. The Cheapest One That Works Costs 56× Less Than Claude Opus.

A real cost-vs-accuracy benchmark of DeepSeek, GLM, Kimi, Qwen, Sonnet, Opus, and GPT-5 for indexing codebases. Why per-file LLM analysis isn't actually expensive if you pick the right model.

Apr 1, 2026 Claude rate limits open source LLMs AI copilot alternative

Claude Is Rate-Limiting Everyone. Here's Why Good Context Beats a Smarter Model.

Anthropic just throttled Claude Opus and Sonnet during peak hours. Developers are canceling subscriptions and looking for alternatives. Here's the argument: a good open-source model with great context beats a frontier model that won't let you use it.

Mar 31, 2026 context rot AI degradation lost in the middle

Why Your AI Gets Dumber the Longer You Talk to It

You're not imagining it. Research tested 18 frontier models and every single one degrades as the conversation gets longer. Here's the science behind it.

Mar 31, 2026 NIAH RULER MRCR

Benchmarking Context Windows: NIAH, RULER, MRCR, and What They Actually Measure

Needle in a Haystack tests one thing. RULER tests another. MRCR tests yet another. Here's what each benchmark actually measures and what it misses.

Mar 31, 2026 API costs token pricing multi-turn cost

The Cost Curve: How Context Length Affects Your API Bill (With Formulas)

A 20-turn conversation costs quadratically more than you'd expect. Here are the exact formulas for multi-turn conversation costs across providers.

Mar 31, 2026 context window AI basics LLM fundamentals

What Is a Context Window? The Simple Explanation Everyone Gets Wrong

The context window is your AI's short-term memory. It's not storage — it's a fixed-size desk. Everything the AI is thinking about right now must fit on this desk.

Mar 31, 2026 training data context window parametric memory

Context Window vs. Training Data: Two Completely Different Things

Training data is everything the model learned. The context window is what it's thinking about right now. Here's why the distinction matters.

Mar 31, 2026 AI coding tool comparison Claude Code vs Cursor vs Copilot AI coding agent problems

Every AI Coding Tool Has the Same Problem. None of Them Will Tell You.

Claude Code, Cursor, Copilot, Codex — they all share the same fundamental flaw: no persistent memory, brute-force file reading, and context that fills up and gets thrown away. Here's what none of them admit.

Mar 31, 2026 Flash Attention GPU optimization SRAM

Flash Attention: How One Paper Made Long Context Possible

Flash Attention never materializes the full n×n attention matrix. Instead, it computes in tiles using fast GPU SRAM. Here's how it works and why it's 2-4× faster.

Mar 31, 2026 future of AI state space models Mamba

The Future: Will Context Windows Grow Forever? (Ring Attention, SSMs, Retrieval-Augmented Everything)

Three competing paradigms: grow context via hardware, replace attention with O(n) alternatives like Mamba, or build external memory systems. Which will win?