강해수

강해수

Jun 19

I let Claude Code run --dangerously-skip-permissions on my production DB. Here's what I changed.

#ai #aiagents #mcp #cloudflare

2

2 min read

강해수

Jun 19

I tracked every async writing block for 90 days. Output nearly doubled — but not how I expected.

#productivity #ai #workflow #knowledgework

2 min read

강해수

Jun 19

The NOT NULL trap that killed my Cloudflare Worker at 2:47am

#cloudflare #serverless #webdev #tutorial

2 min read

강해수

Jun 19

My second brain cost me 140+ hours before it saved me anything

#productivity #ai #workflow #knowledgework

2 min read

강해수

Jun 19

Naver Brand Search shows 5 –9 LTV:CAC and it's the least useful number on my dashboard

#marketing #d2c #paidads #analytics

2 min read

강해수

Jun 17

More parallel subagents made my pipeline slower. Here's the data.

#ai #aiagents #mcp #cloudflare

1

2 min read

강해수

Jun 17

Your R2 presigned URL has a shorter TTL than you think — here's why

#cloudflare #serverless #webdev #tutorial

1

2 min read

강해수

Jun 17

I tracked every context switch across two businesses for 90 days. The 23-minute recovery rule is misleading.

#productivity #ai #workflow #knowledgework

1

2 min read

강해수

Jun 17

40% of my lookalike rebuilds were unnecessary — here's the signal-triggered system I use instead

#marketing #d2c #paidads #analytics

2 min read

강해수

Jun 17

My agent was 'succeeding' on Slack while silently doing nothing — here's the monitoring stack that caught it

#ai #aiagents #mcp #cloudflare

1

2 min read

강해수

Jun 17

D1 read replicas returned stale data for 6 seconds — here's what I measured and how I designed around it

#cloudflare #serverless #webdev #tutorial

2 min read

강해수

Jun 17

I tracked AI rewrite rates for 12 weeks. Notion AI needed fixing 70% of the time. Claude: 30%.

#productivity #ai #workflow #knowledgework

2 min read

강해수

Jun 17

25–35% of our best customers were aging out of retargeting before we ever showed them an ad

#marketing #d2c #paidads #analytics

2 min read

강해수

Jun 15

I cut my Claude system prompt from 4,200 to 820 tokens to save money. It broke production in 3 days.

#ai #aiagents #mcp #cloudflare

2 min read

강해수

Jun 15

My weekly review clocked 14 minutes median — here's the one structural change that made it stick

#productivity #ai #workflow #knowledgework

1

2 min read

강해수

Jun 15

A Merchant Center disapproval wiped 40% of our SKUs the day a 6-week promo launched

#marketing #d2c #paidads #analytics

2 min read

강해수

Jun 15

I wasted $43 rebuilding a Vectorize index the wrong way — here's the $5.50 fix

#ai #aiagents #mcp #cloudflare

2 min read

강해수

Jun 15

I tracked 34 'dead opens' in Obsidian over 3 weeks. That's why I switched to Tana.

#productivity #ai #workflow #knowledgework

2 min read

강해수

Jun 15

Naver catalog quality score: the field that tanked impression share wasn't the one we were optimizing

#marketing #d2c #paidads #analytics

2 min read

강해수

Jun 13

A Durable Object I thought was idle cost me $40/mo — here's exactly why

#cloudflare #serverless #webdev #tutorial

2 min read

DEV Community

Badges

Writing Debut

I let Claude Code run --dangerously-skip-permissions on my production DB. Here's what I changed.

I tracked every async writing block for 90 days. Output nearly doubled — but not how I expected.

The NOT NULL trap that killed my Cloudflare Worker at 2:47am

My second brain cost me 140+ hours before it saved me anything

Naver Brand Search shows 5 –9 LTV:CAC and it's the least useful number on my dashboard

More parallel subagents made my pipeline slower. Here's the data.

Your R2 presigned URL has a shorter TTL than you think — here's why

I tracked every context switch across two businesses for 90 days. The 23-minute recovery rule is misleading.

40% of my lookalike rebuilds were unnecessary — here's the signal-triggered system I use instead

My agent was 'succeeding' on Slack while silently doing nothing — here's the monitoring stack that caught it

D1 read replicas returned stale data for 6 seconds — here's what I measured and how I designed around it

I tracked AI rewrite rates for 12 weeks. Notion AI needed fixing 70% of the time. Claude: 30%.

25–35% of our best customers were aging out of retargeting before we ever showed them an ad

I cut my Claude system prompt from 4,200 to 820 tokens to save money. It broke production in 3 days.

My weekly review clocked 14 minutes median — here's the one structural change that made it stick

A Merchant Center disapproval wiped 40% of our SKUs the day a 6-week promo launched

I wasted $43 rebuilding a Vectorize index the wrong way — here's the $5.50 fix

I tracked 34 'dead opens' in Obsidian over 3 weeks. That's why I switched to Tana.

Naver catalog quality score: the field that tanked impression share wasn't the one we were optimizing

A Durable Object I thought was idle cost me $40/mo — here's exactly why