Erik Dunteman
@erikdunteman
Followers
7K
Following
17K
Media
641
Statuses
7K
LLM response caching @ButterDev_, prev serverless GPUs @BananaDev_ @modal. I like running and drinking water (no ice).
SF
Joined June 2017
We're hiring a Systems Engineer at Butter! Join us in SF to build the muscle memory engine for AI applications. Full Time :: $150k-$250k :: 1-2% Equity https://t.co/bh17hvj0Tb
docs.butter.dev
16
20
227
today no
sf is lit as fuck today
0
0
1
The current “For You” feed is just full of vague-bait. Quote tweets saying shit like “we all know” Reply section full of people trying to deduce wtf they’re supposed to know.
2
0
9
Proud to support some of most ambitious founders in the AI agent ecosystem (biggest perk building in SF) Raising? Want to LP? DM’s open @firecrawl @mem0ai @composio @markokraemer @erikdunteman @MaxMinsker @seanob @TextQL @EyalToledano @agentmail @hazhubble @realshcallaway
it's been an incredible year for agents, and i've had a blast running @theagentfund this year with @adamsilverman & @alexreibman we've made 21 investment across the ecosystem and did a little write up on our learnings, etc.
9
7
39
it's been an incredible year for agents, and i've had a blast running @theagentfund this year with @adamsilverman & @alexreibman we've made 21 investment across the ecosystem and did a little write up on our learnings, etc.
11
12
123
Today I'm excited to share that Nuon is officially open source! You can read more about our announcement here:
nuon.co
Delivering on our mission to make BYOC a standard way to deploy software.
4
10
46
really sick, great move on CUA's part
Today we're announcing cua-bench: a framework for benchmarking, training data, and RL environments for computer-use AI agents. Why? Current agents show 10x variance across minor UI changes. Here's how we're fixing it.
0
1
12
The agent codebase at Butter has transitioned to full Typescript! Here's our current stack: - @bunjavascript for runtime & tests - @e2b for TS sandboxing - @braintrust for evals - @vercel_dev AI SDK for structured generation - @ButterDev_ for caching tests and evals
7
6
90
AI workflow recently has been writing draft one without any AI assist, but then doing no-logical-change refactors with agents to massage out the weird structural decisions made in draft one.
2
0
1
All said positively. It is a genuine gift to have tall mountains to climb.
0
0
2
I've become lead bullets pilled. Anything worth doing is fucking hard.
1
0
11
I'm sorry sir, we don't carry silver bullets in this store. Might I interest you instead in a thousand lead bullets?
0
0
2
Don't think of LLMs as entities but as simulators. For example, when exploring a topic, don't ask: "What do you think about xyz"? There is no "you". Next time try: "What would be a good group of people to explore xyz? What would they say?" The LLM can channel/simulate many
1K
3K
27K
Me when investors ask me about Modal’s business plan
18
25
624
We present @ButterDev_ Changelog #0008 This week: - cookwithbutter . com - data engine rewrite (WIP) - automatic template induction (first POC) https://t.co/hOiyivUHmK
blog.butter.dev
Butter changelog: Thanksgiving recipes at cookwithbutter.com, data engine upgrades, automatic template induction.
2
0
12