Oct 14, 1944. Japan announces a stunning naval victory: 11 American aircraft carriers sunk. They hadn't sunk a single one.
A week before the invasion of the Philippines, Admiral William Halsey's carrier task force hit the Japanese air bases on Formosa, today's Taiwan. The goal was to wipe out Japanese air power before the landings.
Japan threw everything it had at the American fleet. Hundreds of planes, day and night, including an elite force of radar-equipped night torpedo bombers.
They were slaughtered. American fighters and anti-aircraft fire shot down somewhere around 500 Japanese planes over a few days.
But the Japanese got through twice. On Oct 13 a torpedo hit the heavy cruiser USS Canberra. On Oct 14 another hit the light cruiser USS Houston. Both ships were crippled and dead in the water.
American sailors took them under tow and started a slow, painful crawl away from Formosa. Sailors nicknamed the little group Cripple Division One.
Meanwhile, Japanese pilots returning from the night attacks saw fires and explosions on the ocean and reported massive success. Every flash became a sunken carrier.
Tokyo announced it had sunk 11 carriers, 2 battleships and 3 cruisers. Japan celebrated a great victory. The Japanese high command believed it and started planning around it.
Halsey saw his chance. He left the crippled cruisers as bait, hoping the Japanese fleet would come out to finish off the survivors. A Japanese cruiser force did head out. Then its scout planes spotted the entire American fleet, very much afloat, and it turned back fast.
Halsey sent a message to Admiral Nimitz: "All Third Fleet ships reported by Radio Tokyo as sunk have now been salvaged and are retiring at high speed toward the enemy."
Houston was torpedoed a second time. She still stayed afloat. Both cruisers made it all the way to safety.
The false victory cost Japan dearly. Days later, believing the American carrier fleet was shattered, Japan committed to fighting for the Philippines at Leyte Gulf.
That turned into the largest naval battle in history, and it destroyed what was left of the Japanese Navy.
THIS IS A PURE F*CKING GOLDMINE
Anthropic just shared one of the best tricks for Claude Code.
The idea: let Opus 5.5 write the plan as a script, and let a swarm of cheaper agents run it.
Bigger jobs and way fewer Opus tokens, because the script holds the plan and Opus only sees the final answer.
You just launch Claude Code like this:
claude --model opus --effort ultracode
here's the split:
→ Opus 5.5 turns your task into a workflow script
→ Haiku 5.5 agents scan every file in parallel
→ Sonnet 5.5 agents patch each fix in its own isolated copy
→ you stop burning Opus context on a hundred agent transcripts
Opus only shows up at three moments:
→ writing the workflow before the run
→ reading the verified results after it
→ signing off before anything ships
Opus plans. Haiku scans. Sonnet ships.
JEV engineering is the same move one layer down: the forks that need no thinker (which file, which tool, retry or stop) go to JEV in milliseconds, and the big models only see the ones that split
the full tree
> Opus 5.5 runs the main session and writes the workflow
> scan audits every file on Haiku 5.5
> edit patches each fix on Sonnet 5.5 in an isolated copy
> verify cross-checks every finding on Sonnet 5.5
> Opus 5.5 reads only what survived verification
> JEV routes micro-forks in milliseconds
paste the tree and this prompt into Claude Code ↓
"Rebuild my Claude Code setup around this tree:
1. Run the main session on Opus 5.5 with ultracode on.
2. In every workflow, run read-only scan stages on Haiku 5.5 and edit stages on Sonnet 5.5, each edit in its own isolated copy.
3. Add one rule to CLAUDE.md: every finding gets an independent verifier agent before it's reported, and fix loops stop after two rounds with no progress.
4. Route fast deterministic forks (tool choice, path selection, retries) through JEV.
Show me every change as a diff first. No edits until I say go."
↳ https://t.co/1MiKDqGANA
save it, it'll come in handy.
New on our Engineering blog: Creating in-house source control.
In May, we merged around 2,500 code changes per day. In September, we merged more than 30,000 per day. As a result, we became bottlenecked by developer infrastructure built for human (rather than agent) scale.
Here, we share how we built a new version control system—and fully migrated our monorepo from git—in five months: https://t.co/wzNkwHs7IN