design and ai @tryramp, startup advisor, generative art enthusiast, enjoyoor of mornings @earlydayapp, pizza chef dj/producer 🤙
𝖘𝖆𝖓 𝖋𝖗𝖆𝖓𝖈𝖎𝖘𝖈𝖔
Joined June 2009
- Tweets16K
- Following5.1K
- Followers3.5K
- Likes30K
Pinned Tweet
Personal update: after an amazing 6.5 years at Amplitude, I'm dropping in at @tryramp !
Made something fun to celebrate 🛹
Link : playground.willnewton.dev/dr…
Current high score: 24,110
FINALLY broke 30k. You can do it! Don't be afraid!!
More public link with replays : skate.williamnewbton.workers…
And open source code :
github.com/williamnewton/dro…
Personal update: after an amazing 6.5 years at Amplitude, I'm dropping in at @tryramp !
Made something fun to celebrate 🛹
Link : playground.willnewton.dev/dr…
Current high score: 24,110
Dads: Stretching your kids comfort zone is absolutely peak experience. Be a fun dad. 🤘
This quoted post is unavailable.
Weekends are for fun with music and code. Kids love visualizers!
playground.willnewton.dev/p/…
excited to talk about AI x Design tonight at @Amplitude_HQ ! should be fun.
❯ Good morning Claude. It is time to cook. I have a selected frame here in @paper I want you to divergently explore a broad range of at least a dozen visual treatments. Go ahead and use subagents if it is preferred for you to execute here. I want you to create an alphanumeric string that is random and then use that string to influence your direction with colors, typography, spacing, and highlighting to explore a broad visual range of divergent possibilities for how to make our agent-native Amplitude app look. You may bend the rules of the design guideliens in your context, but keep the general blue/white/black/grayscale as primary elements. Do a broad visual explorations exploring conceptual range.
Building an agent is an iterative process that takes a ton of human judgment and refinement over time.
If you don’t understand how your agents are responding across everything your users do, you literally can’t make an agent that’s seriously useful.
Now that building agents has become more mainstream, I’m excited for the explosion of productivity and creativity we’re about to see. But it all starts with knowing if your agents are working in the real world :)
Today we're launching Agent Analytics
Every team shipping an AI agent has the same blind spot. Offline evals pass, you ship, and then you have no idea what's happening in production. AI fails silently. Users ask a question and get different answers. They all look 'engaged' in a classic dashboard. You don't know who got a great response and who got a terrible one.
Agent Analytics solves it:
- Every session scored out of the box on task completion, response quality, friction, safety, and negative feedback
- Topic clustering across thousands of conversations, so you know if a failure hits 1 user or 10,000
- Eval agents that watch for regressions, and if you want will file a Linear ticket or the pull request themselves
- Agent quality sits next to product data, so 'payment scheduling fails 31%' becomes 'which renewals did that cost us?'
The Economist got their agent to a 96.9% task success rate and cut weekly failures 84%.
Included on every plan. Free tier included. amplitude.com/agent-analytic…
Like dialkit but for physical space
What if changing product types could happen directly on the product page?
The lamp transitions between 2 types, which also changes the context in real time. The lights appear to reveal its dimensions.
The whole interactive scene is built in Rive and integrated with Shopify through Konfigur, making it possible to bring this kind of experience to any Shopify store