Tag feed

#prompt-caching

9 posts0 followers

Explore Hashnode

Alternatives

Trending tags this week

Rrecca0120recca0120.hashnode.devMay 9 · 9 min read

36 Days of Claude Code Logs: Silent Model Switching, 11.5x Efficiency Gap

Originally published at recca0120.github.io [The first post]({{< ref "/post/claude-code-cache-ttl-audit" >}}) scanned 95 days of logs and found sub-agent cache TTL silently dropped to 5m. [The second]({{< ref "/post/claude-code-cache-ttl-17-days" >}}...

0

Rrecca0120recca0120.hashnode.devApr 26 · 8 min read

I Audited 3 Months of Claude Code Billing — Most Community Cost-Saving Tips Don''t Work

Originally published at recca0120.github.io This past week, chasing a vague "quota burns faster lately" feeling, I scanned three months of my own Claude Code logs. ~\$127K equivalent cost, 127K turns, four models, hundreds of sessions. The uncomforta...

0

Rrecca0120recca0120.hashnode.devApr 25 · 7 min read

12 More Days Scanned: Claude Code Sub-Agent Cache TTL Has Been 100% 5m for 17 Straight Days — This Isn''t a Regression, It''s the New Default

Originally published at recca0120.github.io [Two weeks ago]({{< ref "/post/claude-code-cache-ttl-audit" >}}) I scanned 95 days of Claude Code logs and found that since 4/9, sub-agents had been 100% downgraded to 5m TTL — 5 consecutive days, 4,840 API...

0

ATAlair Tavares Jralair.hashnode.devApr 17 · 7 min read

Scaling Myself: Processing 26.8 Billion Tokens in 46 Days with Claude Code

For 46 straight days, I submerged myself in code, treating AI not as a glorified chatbot, but as the underlying operating system of my development workflow. Over this non-stop span, I used Claude Code to push my productivity to levels that would conv...

0

FFetchLogicfetchlogic.hashnode.devApr 13 · 8 min read

Anthropic's Silent Cache Cut: What the TTL Downgrade Reveals About the Real Cost of API Dependence

The change appeared in no changelog. No deprecation notice. No migration guide. On March 6th, developers began noticing that their Anthropic cache TTL had silently dropped — from sixty minutes to five. A twelve-fold reduction in cache persistence, im...

0

Rrecca0120recca0120.hashnode.devApr 13 · 7 min read

I Scanned 95 Days of My Claude Code Logs and Found Anthropic''s Second Silent Cache TTL Regression

Originally published at recca0120.github.io The previous post covered prompt caching cost mechanics. While researching it I bumped into a dramatic controversy — in March 2026 Anthropic silently changed Claude Code's cache TTL from 1 hour back to 5 mi...

0

Rrecca0120recca0120.hashnode.devApr 13 · 8 min read

Does a Long Claude Code Session Waste Tokens? A Cost Model Most People Get Wrong

Originally published at recca0120.github.io A common intuition among developers: Claude Code sessions get expensive over time. Context keeps accumulating, every turn resends the entire history, and token costs add up linearly. The obvious conclusion:...

0

TTheDaveAyothedaveayo.hashnode.devMar 27 · 3 min read

How I Cut My Claude Costs Dramatically and Completely Eliminated Session Limits – And How You Can Too

I got fed up. Every time I was deep into a conversation with Claude, the screen would suddenly say “Session limit reached” and kick me out. Even though I had credits. Even though I was paying. It felt

0

SRShani Riversshanirivers.hashnode.devDec 23, 2025 · 7 min read

From Marketing Webinars to a Micro SaaS: The Tool is Live

I love creating content… well, blog posts and YouTube videos that is. My long-form content is usually tied to something I’m building or learning. Whether it’s a YouTube video about learning new frameworks or a coding project I’m currently working thr...

0

#prompt-caching

Search Hashnode

#prompt-caching

Explore Hashnode

Trending tags this week

36 Days of Claude Code Logs: Silent Model Switching, 11.5x Efficiency Gap

I Audited 3 Months of Claude Code Billing — Most Community Cost-Saving Tips Don''t Work

12 More Days Scanned: Claude Code Sub-Agent Cache TTL Has Been 100% 5m for 17 Straight Days — This Isn''t a Regression, It''s the New Default

Scaling Myself: Processing 26.8 Billion Tokens in 46 Days with Claude Code

Anthropic's Silent Cache Cut: What the TTL Downgrade Reveals About the Real Cost of API Dependence

I Scanned 95 Days of My Claude Code Logs and Found Anthropic''s Second Silent Cache TTL Regression

Does a Long Claude Code Session Waste Tokens? A Cost Model Most People Get Wrong

How I Cut My Claude Costs Dramatically and Completely Eliminated Session Limits – And How You Can Too

From Marketing Webinars to a Micro SaaS: The Tool is Live