
Principal Product Manager @Adobe . Tweets are my own.
San Francisco, CA
Joined June 2008
- Tweets7.9K
- Following1.8K
- Followers1.2K
- Likes95
Brian Au retweeted
Serve accuracy and spot serving have rightly gotten more attention this year but you can go deeper by looking individually at the four main spots (AD Wide, AD T, Deuce T, Deuce Wide) players are aiming for. There's a lot of information here but a few high level things:
1. Sinner is lethally accurate in all 4 spots. His four spots land within 5 cm of each other (46–51 cm from the line) and he's at or close to the top in all four zones
2. Fritz's accuracy on T serves on the AD side is incredible at 40cm avg distance from the T line (I've only seen Nakashima post tighter numbers at 39.7cm in same spot)
3. This is a perfect illustration of why Jodar has significant room for improvement as he matures, especially because the rest of his game is already great and serves are a closed skill and more improvable. As far as precedent for serve optimisation goes, for a player his height, he should have room to be both more aggressive, accurate and keep landing a high % of serves in play. A Jodar with a meaningfully higher number of free points on serve should scare the top of the game
4. Alcaraz gets shockingly few free points on his wide serve on the AD side (mentioned before), just 27.7% are unreturned in that spot (for context his biggest rival Sinner is just under 50%). He's 2nd last overall for unreturned just in front of De Minaur. He still wins a high % of 1st serve points because his +1 and rest of his game are so elite, but in certain matchups this is a remaining vulnerability (especially because that serve will face more break points than the deuce side)
5. The top 5 guys for placement, power, and risk (first serve in %) are all 6ft3-6ft6. Their height and technique mean they often don't have to trade-off much between power, accuracy, and serve in%. In comparison a player like Tiafoe has to gamble more, his best serve unreturned wise (47%) is also his riskiest (just 44% in)
6. Shelton's lefty T serve on the AD side is essentially unplayable (62% unreturned, 137mph avg speed, >60% in). Part of this is a second order effect of the threat of his lefty slider out wide on the AD side: you'll see returners having to hedge their positioning a bit wider than usual, leaving space open for this T serve instead. He goes 50% wide and 40% T (rest body) on these serves so opponents are kept guessing
7. The 4th image measures 'deviation' which is showing how much the ball moves from a straight path on 1st serves. This measures the 'swing' in serves basically, so a higher degree out wide on the deuce side would be a slice serve swinging away from the returner, away from the court. Alcaraz and Fritz both have really swingy deuce wide serves, and this is part of why Alcaraz's best performing serve by far unreturned wise (44%) is that wide serve on the deuce side
Brian Au retweeted
i recently added this command to the claude-api skill. run it in Claude Code to fix common prompting "anti-patterns" that can hobble frontier models:
/claude-api prompt-audit
patterns include:
1/Verification rituals. Instructions like "double-check your work” or "verify twice before responding” are often taken literally by frontier models and can waste tokens.
2/ Thoroughness and emphasis boosters. "Be maximally thorough," "CRITICAL: YOU MUST ALWAYS…" can lead to verbosity and extra tool calls when working with frontier models.
3/ Mandatory procedures and scratchpad scaffolds. Fixed step processes (e.g., "think step by step in a scratchpad") or reasoning templates are rituals that frontier models don't need. This scaffolding can stack on top of native reasoning and use unnecessary tokens.
4/ Stale examples. Few-shot examples tuned to an older model's failure modes can teach a frontier model to imitate long reasoning chains on requests that don't need them.
5/ Contradictory rules. Frontier models are better at instruction following. Contradictory instructions ("always refund within policy" vs. "never issue refunds without escalation") can be followed more literally by frontier models, resulting in degraded performance.
6/ Dated configuration. Settings written for an older Claude generation (e.g., manual thinking budgets) can be rejected by the Claude Platform with newer models.
these patterns accumulate in prompts over time, and can quietly degrade performance when upgrading to newer models. a common reason is the frontier models are better at instruction following, so these anti-patterns steer them to spend unnecessary tokens.
example: i tested a migration from Opus 4.8 to Opus 5 on an internal customer support benchmark. with Opus 5 (and other frontier models like Fable 5.1), verification rituals ("verify twice") use unnecessary tokens by duplicating work. emphasis boosters ("be maximally thorough") become dozens of unneeded searches.
applying prompt audits can improve performance and reduce cost (as shown in example attached and will be sharing a full write-up soon).
also, the skill is also open source and some of this guidance likely applies generally across frontier models
github.com/anthropics/skills…
xitter.gq/petergyang/status/2094…
Brian Au retweeted
This is so interesting because there's a measurable impact from newer/older balls on rates of aces, holds, unreturned, double faults, 1st serve in %, rally length et al. So if balls are changed more often, in theory things like serve dominance may increase (from existing all time highs btw). Impact will likely be small (double faults also rise with newer balls, 1st serve in % decreases modestly, & hold rate changes are smaller). Probably only detectable for season-aggregated unreturned/ace rates and multi-season aggregates for hold rates. But it's probably easy to underestimate 2nd & 3rd order effects supposedly small tweaks like this can have on the way the game is played. Tennis is already massively short point and serve dominant, and changes like this nudge it slightly further in that direction
Brian Au retweeted
The thing about people talking about building with AI is that they always talk about how they’re building, what tools they use, and how much they use them.
Much less is said about what they actually built, or what impact it had.
The tool becomes the job. And the purpose.
Brian Au retweeted
The obvious path for designers in the AI age is to move closer to code.
But the more valuable path may be upstream: closer to the customer, the business, and the problem.
If everyone can prompt agents to code, the scarce skill becomes knowing why, what, and how to build.
Brian Au retweeted
"There's a good looking guy with white hair looking down on us right now that's happy!" 🥹
Tommy Lloyd pays homage to the late Lute Olson in his postgame interview 🙏
#MarchMadness
Brian Au retweeted
For my friends who are still using UV and might be a little weary about recent compromises to PyPi packages, stick this in your pyproject.toml.
You can let all of those pip users find and report the compromises...
We're excited to announce duckdb-skills, a DuckDB plugin for Claude Code!
We think the embedded nature of DuckDB makes it a perfect companion for Claude in your local workflows.
The skills supported include:
+ read-file and query – uses DuckDB's CLI to query data locally, unlocking easy access to any file that DuckDB can read.
+ read-memories – a clever idea to store your Claude memories in DuckDB and query them at blazing speed.
These are powered by two additional skills:
+ attach-db – gives Claude a mechanism to manage DuckDB state through a .sql file linked to your project.
+ duckdb-docs – uses a remote DuckDB full-text search database to query the DuckDB docs and answer all of your (and Claude's own) questions.
github.com/duckdb/duckdb-ski…
Brian Au retweeted
Replying to @thdxr
yeah it is but everything in moderation. Internally we always talked about main quest and side quests.
Everyone should focus on the main quest, and moderately or not all on side quests.
Both quest lines feel productive but only one of them advances the main mission of the company.
Brian Au retweeted
Introducing the Google Workspace CLI: github.com/googleworkspace/c… - built for humans and agents.
Google Drive, Gmail, Calendar, and every Workspace API. 40+ agent skills included.
Brian Au retweeted
In the next version of Claude Code..
We're introducing two new Skills: /simplify and /batch. I have been using both daily, and am excited to share them with everyone.
Combined, these kills automate much of the work it used to take to (1) shepherd a pull request to production and (2) perform straightforward, parallelizable code migrations.



























