Partial day — headline is the last complete day (2026-10-13).
twitter
tracked since 2026-10-03T10:40:24.743954+00:00
conclusion
Measured mentions per day
day (UTC)
mentions
2026-09-27
0
2026-09-28
0
2026-09-29
0
2026-09-30
1
2026-10-01
5
2026-10-02
16
2026-10-03
7
2026-10-04
11
2026-10-05
14
2026-10-06
12
2026-10-07
4
2026-10-08
7
2026-10-09
7
2026-10-10
7
2026-10-11
0
2026-10-12
8
2026-10-13
4
2026-10-14
0
derived from
jev
word-boundary match over title + text, query='timeline' · window
2026-09-27 → 2026-10-15
· 103 matching posts
evidenceShowing 20 newest of 103
•
twitterhayatomaruu9hneutral
Anthropic engineer:
"You're not supposed to prompt Opus 5.5
You're supposed to build a system that prompts itself. At Anthropic, 90% of our engineers have already built such systems"
In 30 minutes she shows exactly how Anthropic builds agents that remember, fix their own mistakes and get smarter with every run.
Thi…Anthropic engineer:
"You're not supposed to prompt Opus 5.5
You're supposed to build a system that prompts itself. At Anthropic, 90% of our engineers have already built such systems"
In 30 minutes she shows exactly how Anthropic builds agents that remember, fix their own mistakes and get smarter with every run.
This beats any $1000-$2000 course on agents I've seen.
Watch it, then read the full guide for Opus 5.5 loop + jev engineering.
THIS IS F*CKING BRILLIANT. Anthropic’s docs describe one of the smartest ways to run Opus
three models work on one task and each does only the part it’s cheapest at. start Claude Code like this:
claude -> model sonnet -> advisor opus
you stop paying Opus prices for work that doesn’t need Opus
> Sonnet runs the sess…THIS IS F*CKING BRILLIANT. Anthropic’s docs describe one of the smartest ways to run Opus
three models work on one task and each does only the part it’s cheapest at. start Claude Code like this:
claude -> model sonnet -> advisor opus
you stop paying Opus prices for work that doesn’t need Opus
> Sonnet runs the session and edits the code
> Haiku 5.5 searches the repo and docs, at up to 40x less per token than Opus 5.5
> Opus waits as the advisor and reads the whole session, tool calls included
Claude usually calls Opus at three points:
> before it locks a plan
> when the same error comes back
> right before it says it’s done
Sonnet drives. Haiku scouts. Opus signs off
and if you go one step lower, most of an agent’s day is tiny choices like picking a file or a tool, or deciding whether to retry. Jev takes those first, so Sonnet and Opus only see the moments that are really open
the whole setup
Sonnet runs the main session
Explore on Haiku 5.5 searches and pulls docs, read-only
Opus on call as the advisor
give Claude Code this setup and the prompt below ↓
“Set up my Claude Code exactly like this:
Main model Sonnet, advisor Opus, saved as advisorModel in my settings.
One read-only subagent named exactly Explore, on claude-haiku-5-5, to replace the built-in one.
One line in CLAUDE.md: ask the advisor before locking a plan, after the same error twice, and before calling a task done, not more often.
Check what could turn the advisor off (CLAUDE_CODE_DISABLE_ADVISOR_TOOL, DISABLE_TELEMETRY) and only report it.
Show every change as a diff first. Don’t edit anything until I say go.”
https://t.co/xpEE0hFQ6d
before you turn it on:
every advisor call re-reads the whole transcript with no cache. on a 200k session that’s about $0.80 in input, against $0.04 for a normal request. keep it to a few calls
the built-in Explore agent runs on your main model unless a subagent is named exactly Explore. any other name does nothing
turning the advisor on or off keeps y
WEEK 1:
Your agent runs exactly like before.
Jev answers every decision silently.
You log both answers.
WEEK 4:
Jev makes the small calls.
Opus 5.5 only does the hard thinking.
Your bill is a third of what it was.
No risky switch.
No big rewrite.
Just a decision layer, rolled out slowly.
Full plan below 👇WEEK 1:
Your agent runs exactly like before.
Jev answers every decision silently.
You log both answers.
WEEK 4:
Jev makes the small calls.
Opus 5.5 only does the hard thinking.
Your bill is a third of what it was.
No risky switch.
No big rewrite.
Just a decision layer, rolled out slowly.
Full plan below 👇
This is a playbook worth studying if you’re building with coding agents.
A guide to building a Jev harness breaks down how the right setup can deliver:
→ Up to 200x faster execution
→ Up to 400x lower costs
The model isn’t always the bottleneck. The harness around it can make a massive difference in both speed and co…This is a playbook worth studying if you’re building with coding agents.
A guide to building a Jev harness breaks down how the right setup can deliver:
→ Up to 200x faster execution
→ Up to 400x lower costs
The model isn’t always the bottleneck. The harness around it can make a massive difference in both speed and cost.
Here’s how to put this into practice:
1. Rebuild your setup
Give the PDF and the article below to Claude Code or Codex. Ask it to redesign your agent workflow around Jev.
2. Pick your most expensive workflow
Find the agent task your team spends the most money running. Try rebuilding it in one evening.
3. Show the results
Compare the before-and-after latency, token usage, and cost. Bring the numbers to your manager.
One evening of harness engineering could save your team a serious amount of money and give you a system most teams haven’t even thought to build.
https://t.co/W9XY9ne0X0
THIS IS ACTUAL F*CKING GOLD
Anthropic just dropped one of the best tricks for Claude Code.
The idea: run Sonnet, Haiku and Opus 5.5 together, each one doing only the job it's best at.
Better output and way fewer tokens, because every task lands on the right model.
You just launch Claude Code like this:
claude --mo…THIS IS ACTUAL F*CKING GOLD
Anthropic just dropped one of the best tricks for Claude Code.
The idea: run Sonnet, Haiku and Opus 5.5 together, each one doing only the job it's best at.
Better output and way fewer tokens, because every task lands on the right model.
You just launch Claude Code like this:
claude --model sonnet --advisor opus
here's the split:
→ Sonnet 5 runs the session and writes the code
→ Haiku 5.5 explores files and docs in parallel
→ Opus 5.5 stays on call as advisor for the big calls
→ you stop burning Opus context on work that doesn't need it
Opus only steps in at three moments:
→ before a plan locks
→ when the same error shows up twice
→ before calling it done
Sonnet builds. Haiku explores. Opus reviews.
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
> Sonnet 5 runs the main session and edits code
> search maps the repo on Haiku 5.5
> grep finds call sites on Haiku 5.5
> docs pulls the API reference on Haiku 5.5
> Opus 5.5 on call as advisor, reads the whole transcript when consulted
> 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 Sonnet 5 with Opus 5.5 as advisor.
2. Create three read-only Haiku 5.5 subagents: search (maps the repo), grep (finds call sites), docs (pulls the API reference).
3. Add one rule to CLAUDE.md: consult the advisor before locking a plan, after the same error twice, and before calling a task done.
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/vOifNVjYxB
you'll use this every single day.
A Jev engineer hasn’t waited for CI in 3 months. His code still ships faster and safer than the rest of the team’s.
He didn’t make CI faster. He made every change prove itself against real production traffic.
Here’s the setup:
→ SHADOW: Mirror real requests to the new code. Users never see its responses.
→ JUDGE: C…A Jev engineer hasn’t waited for CI in 3 months. His code still ships faster and safer than the rest of the team’s.
He didn’t make CI faster. He made every change prove itself against real production traffic.
Here’s the setup:
→ SHADOW: Mirror real requests to the new code. Users never see its responses.
→ JUDGE: Compare old and new outputs request by request. Score intent, not just exact matches.
→ SWAP: Ship changes that pass, with a 200ms rollback ready if anything breaks.
Traditional CI tests against assumptions about user behavior.
Shadow traffic tests against what users actually do.
And when something fails, the team gets the exact request that broke it, with the input attached.
They still write tests. They just don’t treat a green checkmark as proof that production will be fine.
Also, if you wanna learn how to build a model like Jev, read the article below. 👇
https://t.co/dkCYR9sm9X
Decision-making models are the new hot thing in the industry.
SubQ Snap is the latest model to come out, and it looks pretty strong:
• Up to 384K tokens of context in a single call
• Processes 100K tokens and returns a decision in <1 second
• Uses sparse attention to skip unnecessary token comparisons
• Handles text …Decision-making models are the new hot thing in the industry.
SubQ Snap is the latest model to come out, and it looks pretty strong:
• Up to 384K tokens of context in a single call
• Processes 100K tokens and returns a decision in <1 second
• Uses sparse attention to skip unnecessary token comparisons
• Handles text and images in the same context
• Trained to handle structured, data-heavy decisions
• Leads Jev on 13 of 16 Decision Index benchmarks
Introducing SubQ Snap, the first long-context decision model for real enterprise work.
Snap is the only decision model or LLM where compute per token stays constant as context grows.
• 384K tokens of multimodal context, 12x larger than Jev.
• 61.7 on Decision Index v0.2.1 vs Jev’s 57.9, outperforming on 13 of 16 ente…Introducing SubQ Snap, the first long-context decision model for real enterprise work.
Snap is the only decision model or LLM where compute per token stays constant as context grows.
• 384K tokens of multimodal context, 12x larger than Jev.
• 61.7 on Decision Index v0.2.1 vs Jev’s 57.9, outperforming on 13 of 16 enterprise benchmarks.
•Processes 100K tokens and returns a decision in under one second.
•$0.05/MM input tokens. Output is free.
Available today to anyone.
The first 10,000 sign-ups get 100M free input tokens.
We’ll likely hit capacity fast and will add users as we add compute. Compute offers welcome 🙂.
Latency will be variable as we tune the platform to this new model and category of users.
Model speeds are 80ms for minimal context, 250ms for 32k, 660ms for 100k, and 2.5 seconds for 384k.
Visit: https://t.co/Nc76iJF2a8
Your coding agent is a surgeon doing the paperwork.
Opus 5.5 is brilliant.
And you're paying it to decide which file to open.
Which test to run.
Whether to retry.
That's not surgery.
That's admin.
Give the admin to Jev.
Keep the surgeon in surgery.
Same results. A third of the bill.
Full architecture below 👇Your coding agent is a surgeon doing the paperwork.
Opus 5.5 is brilliant.
And you're paying it to decide which file to open.
Which test to run.
Whether to retry.
That's not surgery.
That's admin.
Give the admin to Jev.
Keep the surgeon in surgery.
Same results. A third of the bill.
Full architecture below 👇
Do the math on 1,000 coding tasks a month.
Opus 5.5 doing everything:
$6,420
Jev + Opus 5.5:
$2,000
That's $4,420 a month back.
$53,040 a year.
Same tasks.
Same model writing the code.
The only difference is who makes the small decisions.
Full build below. Bookmark this. https://t.co/HYnKk0NfmIDo the math on 1,000 coding tasks a month.
Opus 5.5 doing everything:
$6,420
Jev + Opus 5.5:
$2,000
That's $4,420 a month back.
$53,040 a year.
Same tasks.
Same model writing the code.
The only difference is who makes the small decisions.
Full build below. Bookmark this. https://t.co/HYnKk0NfmI
THE BEST AI AGENT IN 2026 USES 4 MODELS.
Not one.
Opus 5.5 for the hard thinking.
Sonnet 5.5 for the building.
Haiku 5.5 for the volume.
Jev for the yes or no decisions.
Picking which model does which job used to take an engineer.
Now Opus 5.5 does it for you.
BTHE BEST AI AGENT IN 2026 USES 4 MODELS.
Not one.
Opus 5.5 for the hard thinking.
Sonnet 5.5 for the building.
Haiku 5.5 for the volume.
Jev for the yes or no decisions.
Picking which model does which job used to take an engineer.
Now Opus 5.5 does it for you.
B
A Jev engineer hasn’t waited on a CI pipeline in three months, and his changes still reach production faster and safer than the rest of the team’s
He never sped the pipeline up. He routed around it, letting every change prove itself on live traffic users never see.
> SHADOW - each change runs beside production on a m…A Jev engineer hasn’t waited on a CI pipeline in three months, and his changes still reach production faster and safer than the rest of the team’s
He never sped the pipeline up. He routed around it, letting every change prove itself on live traffic users never see.
> SHADOW - each change runs beside production on a mirrored copy of real traffic
> JUDGE - the Jev system compares both outputs request by request, never by averages
> SWAP - a change that matches or beats live behaviour goes in, with a 200 ms rollback armed
CI was built for a world where real traffic was too risky to touch, so it tests against guesses about what users do.
Shadow traffic removes the guess. The new code sees exactly what users send, and its answers simply never reach them.
The comparison is the hard part. Two outputs can differ and both be right, so Jev scores intent instead of exact matches.
Most changes clear in minutes. The ones that fail get caught on the exact request that broke them, input attached.
The armed rollback is what makes it safe. One bad response and the old path is back before the next request lands.
His team still writes tests. They just stopped treating a green checkmark as proof that real users will be fine.
His full setup, from the traffic mirror to the rollback trigger, is written up in the article attached here.
if you've noticed that you get more visual answers within omp recently (like on the left) that's thanks to the Jev powered "auto-graph" pipeline we added btw (detailed on the right)
for $0.0001 you get a much richer response, yet another great use case 🫡 https://t.co/OFPgRYFs9Bif you've noticed that you get more visual answers within omp recently (like on the left) that's thanks to the Jev powered "auto-graph" pipeline we added btw (detailed on the right)
for $0.0001 you get a much richer response, yet another great use case 🫡 https://t.co/OFPgRYFs9B
This is pure gold if you're building coding agents.
Jev founder Diogo Amago just dropped a PDF on building a Jev harness.
Their tests show:
→ Up to 200x faster
→ Up to 400x cheaper
The crazy part?
The model might not be your bottleneck anymore. The harness around it can make a massive difference in both speed and co…This is pure gold if you're building coding agents.
Jev founder Diogo Amago just dropped a PDF on building a Jev harness.
Their tests show:
→ Up to 200x faster
→ Up to 400x cheaper
The crazy part?
The model might not be your bottleneck anymore. The harness around it can make a massive difference in both speed and cost.
Here's how I'd use this to get ahead:
Give the PDF + article to Claude Code or Codex.
Ask it to rebuild its own setup around Jev.
Pick the most expensive agent workflow your team runs.
Benchmark the old vs. new setup and take the cost savings to your manager.
One evening of work could save your team a serious amount of money.
And you'll be the person who figured out how to make your coding agents faster and cheaper while everyone else is still paying the old bill. 👀
Make the benchmark claims more preciseMake the opening hook more specific
https://t.co/W9XY9ne0X0
10 000 décisions SEO te coûtent 42 centimes maintenant
J'ai monté le MEILLEUR agent de maillage interne avec Jev AI
(il décide où pointe chaque lien de ton site pour quelques centimes)
Je te donne le setup complet gratuitement
Comment ça marche :
→ tu lui donnes ton sitemap (ça tourne sur n8n, Make ou un simple sc…10 000 décisions SEO te coûtent 42 centimes maintenant
J'ai monté le MEILLEUR agent de maillage interne avec Jev AI
(il décide où pointe chaque lien de ton site pour quelques centimes)
Je te donne le setup complet gratuitement
Comment ça marche :
→ tu lui donnes ton sitemap (ça tourne sur n8n, Make ou un simple script Python)
→ pour chaque page, il choisit la meilleure cible de lien parmi 255 pages max
→ il te rend son choix avec un score de confiance entre 0 et 1
Il lit le contenu de chaque page, garde en auto les liens au-dessus de ton seuil et t'envoie le reste à vérifier
1. Follow
2. Like + signet
3. Commente "jev"
(guide dispo en français et en anglais)
Around 3 weeks ago introduced Laya, the first open-source version of JEV, and it has reached a very large scale. Today I am open-sourcing a new physics-based typed decision model called Vega.
The interesting parts. It is only 800 million parameters (4B is also there), with 73k token support, image support, and surpassi…Around 3 weeks ago introduced Laya, the first open-source version of JEV, and it has reached a very large scale. Today I am open-sourcing a new physics-based typed decision model called Vega.
The interesting parts. It is only 800 million parameters (4B is also there), with 73k token support, image support, and surpassing many of the jev benchmarks in single-shot. Implemented an engine+adapter as a Test-Time Training Architecture.
Imagine you are giving input to the model like "ignore all previous conversations," which is called the observation, along with a question, "Is this trying to override the instructions of the model?" and your possible outcomes are YES or NO. The whole prompt is passed to an LLM/VLM (yeah, image-supported), then from the hidden layers, we can extract the observation, question, the yes and no, and last token. These will be vectors, and we can project it by multiplying with wieghts; then, first, using the observation projection, we can create a landscape with valleys; then, using the question and last-token vectors, we can initialise a ball in the valley, and using YES or NO, we can get the candidate location of the ball. Depends on the number of outcomes we create valleys; that is, here YES and NO, so 2. When the ball fall on YES valley, we get the answer. There is friction also influencing how the ball moves.
TLDR: It's like creating valleys of outcomes that we need and throwing a ball that will slow down based on friction and settle on the best outcome
Blog: https://t.co/GFaP7djUBQ
GitHub: https://t.co/ai8SnOElQ8
HuggingFace Model Card: https://t.co/6jEWA8eoBG
HuggingFace Space: https://t.co/GvzEKjkWB2
BEFORE:
Opus 5.5 does everything.
Reads every file.
Decides every retry.
$6.42 per task.
AFTER:
Jev makes the small decisions.
Opus only does the hard reasoning.
Confidence thresholds catch the risky calls.
$2.00 per task.
Same agent.
Same code quality.
A third of the bill.
Full build below 👇BEFORE:
Opus 5.5 does everything.
Reads every file.
Decides every retry.
$6.42 per task.
AFTER:
Jev makes the small decisions.
Opus only does the hard reasoning.
Confidence thresholds catch the risky calls.
$2.00 per task.
Same agent.
Same code quality.
A third of the bill.
Full build below 👇
WHOEVER LEAKED THIS CLAUDE CODE SETUP DESERVES A RAISE
I thought the best way to use Claude Code was to just throw Opus at everything.
Apparently not.
Anthropic’s setup is much more interesting:
→ Sonnet 5.5 runs the main session and actually builds the thing
→ Haiku 5.5 swarms the repo for file discovery, AST scan…WHOEVER LEAKED THIS CLAUDE CODE SETUP DESERVES A RAISE
I thought the best way to use Claude Code was to just throw Opus at everything.
Apparently not.
Anthropic’s setup is much more interesting:
→ Sonnet 5.5 runs the main session and actually builds the thing
→ Haiku 5.5 swarms the repo for file discovery, AST scans, grep, and docs
→ Opus 5.5 stays in the background as the architect
→ Opus gets called in before big plans, after repeated test failures, and before shipping
→ JEV handles the boring mechanical routing in ~16ms so frontier models only wake up when real reasoning is needed
so you get:
Opus reviews. Sonnet builds. Haiku swarms. JEV routes.
the interesting part isn’t using multiple models.
it’s giving each model a different job.
Instead of burning Opus context on finding files and running routine commands, you save it for the moments where architectural judgment actually matters.
Plan on high.Delegate on medium. Keep Opus on call.
❤️ 𝐋𝐈𝐊𝐄 ❤️
🔁 𝐍𝐄𝐄𝐃 𝐑𝐓 🔁
💬 𝐂𝐎𝐌𝐌𝐄𝐍𝐓 💬
🔖 𝐃𝐎 𝐁𝐎𝐎𝐊𝐌𝐀𝐑𝐊 🔖
👤 𝐅𝐎𝐋𝐋𝐎𝐖 𝐌𝐄 👤 @fran_alex897
i'm sorry, Jev can WHAT?
it can take your website link, find people who need your product, email them and book the calls?
since when?? i've been doing this by hand for a YEAR.i'm sorry, Jev can WHAT?
it can take your website link, find people who need your product, email them and book the calls?
since when?? i've been doing this by hand for a YEAR.