
top 1% parallel parker, 12th percentile dishwasher loader | @ATLTUG co-lead | Solution Leader @analyticvizion | stats & running
Atlanta, GA
Joined August 2009
- Tweets6.1K
- Following1.3K
- Followers1.8K
- Likes13K
Pinned Tweet
Something doesn't look right...
(It's acute joke.)
#mathjokes
THIS IS A ACTUAL F*CKING GOLDMINE
Anthropic just shared one of the best tricks for Claude Code
The idea: run Sonnet, Haiku 5.5 and Opus 5.5 at once, each doing only what it's best at
Better results and way fewer tokens: every task lands on the right model
You just start Claude Code like this:
CLAUDE_CODE_SUBAGENT_MODEL=haiku claude --model sonnet
here's how it splits:
→ Sonnet 5.5 runs the main session, writes the code, and applies fixes
→ Haiku 5.5 swarms the repo in parallel: search, grep, docs, and test audits at 340 tok/s
→ Opus 5.5 stays on call as architect for the big decisions and signs off on diffs
→ you stop burning Opus context on routine tasks and stop flooding main session with raw logs
Opus only steps in at three moments:
→ before a plan locks
→ when the same error hits twice
→ before calling it done (signing off on git diff)
Sonnet builds, Haiku swarms, 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 5.5 runs the main session and coordinates code implementation
> search maps the repo on Haiku 5.5 (read-only, 340 tok/s)
> grep finds call sites on Haiku 5.5 (0.05x cost, isolated context)
> audit scans diffs in a temporary worktree (isolation: worktree)
> Opus 5.5 on call as architect, reviews synthesized diff before sign-off
> JEV routes micro-forks in milliseconds
paste the tree and this prompt into Claude Code ↓
"Reconfigure my Claude Code setup around this tree:
1. Run the main session on Sonnet 5.5 with Opus 5.5 as architect
2. Create three custom Haiku 5.5 subagents in .claude/agents/: search (maps repo), grep (finds call sites), and audit (runs isolated diff checks in a worktree)
3. Set isolation: worktree on audit so modifications never touch my working branch
4. Add one rule to CLAUDE.md: delegate all repository searches and test runs to Haiku 5.5 subagents to keep main context under 10k tokens
5. 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"
↳ code.claude.com/docs/en/sub-…
save it, it'll come in handy
Anna Foard retweeted
The Data + Love podcast returns with @stats_ninja
In this episode we discuss whether AI CEOs are stupid or liars, the tradeoffs of enthsiasm, and the importance of tenacity
open.spotify.com/episode/3l3…
Anna Foard retweeted
We'll be spending a lot more time trying to understand the outputs of language models. A few thoughts, tips & tricks:
Writing. Something I've had success with: Ask your LLM to explain something in ASD-STE100, it's a controlled language specification originally developed for aerospace maintenance documentation. LLMs well-versed in this language and it comes with heavy constraints on clean writing style that I often find a lot more readable. Sometimes I've tried to soften it a bit e.g. ask for "80% of the way to ASD-STE100" because the spec is quite stringent. But even better:
Diagrams / images. Instead of writing, ask your LLM to create a diagram. These can be a lot easier to process, parse, and understand. But even better:
Web pages. Ask for output "in HTML" to get a beautiful, interactive webpage. LLMs are getting really good at frontend and can create beautiful experiences, animations, etc. But even better:
Explainer videos. The output format I am most bullish on is fully custom / bespoke explainer videos generated on any arbitrary topic. Experiment with things like "Create a 3b1b style video explainer on X. Use my ElevenLabs API key for audio narration". (you'd need an API key for the latter or you can ask your LLM to find you decent free alternatives that use your local compute). This is actually starting to work!
In summary:
- As LLMs get better, they will do more and more of the legwork autonomously, and a lot more of our work will rise up the abstractions into oversight and understanding.
- Luckily, LLMs can help here too because as intelligence and code are increasingly abundant, you can ask for large, custom, discardable software artifacts (e.g. web apps, video explainers) that would have never made sense to create before. Push the boundaries here and you'll be surprised.
Anna Foard retweeted
The best way to improve is not to focus on results. It’s to focus on habits.
27 studies: Behavior change goals are 3x more effective than outcome goals in boosting performance. Growth depends on altering the actions we can control.
A key to progress is concentrating on process.
Preparing/training for @salesforce #Dreamforce with 15 miles of mountain running and @Metallica 🔥
@Benioff @ryanaytay
#df25
Anna Foard retweeted
Cursing is rarely a symbol of low class. It's often a mark of high authenticity.
Evidence: Swearing predicts higher rates of honesty and integrity. It signals a willingness to prize candor over courtesy.
A little profanity shows that you're being real and you do give a damn.
Causation is "proven" scientifically through randomized controlled experiments. Evidence for causation without a randomized, controlled experiment most often requires a massive amount of data over many years. Controlling for confounding variables in observational data is difficult.
Today's "healthcare" has been built on the backbone of poorly reviewed studies financed by Big Pharma that ignore the importance of correlation when it comes to negative side effects, but plants it's flag when a desired result is found. This is not science.
I've decided to come back to Twitter, eh X. And I'm here for all things statistics, data science, and fitness. Looking for respectful conversations that prioritize critical thinking and the OG Scientific Method.
Hey look, I took a selfie with myself!
My 10-year-old just referred to what most people would call "forecasting" as "future guessing" and I'm going to start calling it that now.
Atlanta Tableau User Group is IN PERSON Thursday 9/12 @ 1PM Use this link to register 👇
Join us as Bradly Davidson presents ARE YOUR LOGISTICS LOGICAL?! Our second block of time will be break-out table discussions grouped by industry vertical.
usergroups.tableau.com/event…
Anna Foard retweeted
There's stil time to get your Ambassador nominations in!
Social Ambassadors connect the #DataFam through digital and in-person experiences.
Learn more about all seven branches of the #TableauAmbassador program, and apply or nominate others for this year's cohort by July 11.
tabsoft.co/3yHdWZI
























