Ben Levine (@blevined) header
Ben Levine (@blevined) profile picture

Helping you w your finances, building ai agents @getcandidly. PhD @Columbia dropout. Formerly @twitter doing data+revenue things.

bay area
Joined March 2010
Photo by Ben Levine (@blevined)Photo by Ben Levine (@blevined)Photo by Ben Levine (@blevined)Photo by Ben Levine (@blevined)Photo by Ben Levine (@blevined)Photo by Ben Levine (@blevined)Photo by Ben Levine (@blevined)Photo by Ben Levine (@blevined)Photo by Ben Levine (@blevined)Photo by Ben Levine (@blevined)
Coding agents let me take on more projects. They also leave more work waiting on us. At Candidly, we run @cursor_ai agents from a shared @NotionHQ board. This has been very helpful and we're clearly in the early innings of a software factory. More below!
3
2
2
21
7,344
Ben Levine retweeted
It is amazing that human language turned out to be so unreasonably effective at modeling the universe that training a system to predict what comes next in a sentence would get us this far.
475
941
172
14,390
399,507
me and a coding agent just fixed our sonos which had been broken for 1+ years ama
2
6
238
someone wrote an entire paper on situationships and game theory & bayesian…
53
449
142
4,159
246,004
first in my bloodline to write a sentence like this one
1
4
155
lets go tweeps!
Top 50 most talent-dense breeding grounds for founders. Palantir at #1, Stanford at #2 and surprisingly Yahoo at #3.
1
2
243
This does it for me.
UI and math time! UI = f(props) is the most famous line in React. But that f is JavaScript, not math. Inspired by @_chenglou’s Handwired project, I compiled a <Dashboard/> component into a real math function, and asked what becomes possible. calculus-of-interfaces.verce…
95
It's very simple: when @sydneyrunkle writes, I read (and usually send it around internally). This is just so clear and accessible about what's going on and why we devote so many calories to evals as a fundamental part of our product development. 👏
6
189
It’s a golden age to be a programmer. Models are capable of pretty incredible things but only in the hands of those with the domain knowledge and expertise to steer them. They’re not smart enough to come up with novel/optimal solutions to difficult tasks out of the distribution, but will usually do very well if you give them a plan at the appropriate level of abstraction. Models being good enough to be directed but not yet good enough to self direct represents an insane (and ephemeral) advantage for experts. It makes the fabled 10x programmer into the 1000x programmer that absolutely runs rings around everyone else. The value is no longer implementing the solution, it’s knowing what solution to implement.
2
1
17
689
Ben Levine retweeted
I've managed fully remote engineering teams for ~7 years, from small teams to rapidly growing orgs spanning 10+ countries. I've developed some pretty strong opinions about what makes remote work actually work: <rant> 1. YOU MUST GET "THE TEAM" TOGETHER ONCE A QUARTER, NO EXCEPTIONS. The faster the team/company is growing, the more critical this is. If you follow no other piece of advice follow this one. Remote work is built on a foundation of trust which for most humans can only be built in person. Think about a person you know well who has extremely dry humor. Would you be able to detect that characteristic if you met them for the first time over SMS or slack? Bonds are built over dinner, a beer, an in person hackathon. And those bonds carry forward to the rest of the time spent remote. **for some definition of "the team" - can be a mix of immediate team, org, company-wide, all are useful at different cadences. 2. Managers must be pros and must be highly technical. Many non-technical managers rely on lazy half-ass signals like "who left the office last". As a remote manager you literally have no option other than to inspect the concrete outputs to understand what is going on. This tip is also very useful to non-remote teams 🙂 3. No juniors except for exception cases. Exceptional cases are obviously exceptional. Gotta build the muscle before you play on hard mode - for mere mortals at least. Don't expect someone to learn "how to remote" and "how to engineer" simultaneously. It is too much. 4. Establish a physical center of gravity, ideally 20-30% of staff. Company can still be fully remote and have "an office" where company events are held, some fraction of local employees cowork once a week even if on different teams. 5. Regardless of critical mass, all meetings must be remote first. None of this conference room with 10 people dialing into a zoom meeting with 3 other people nonsense. 100ms latency of conversation is real and you are at a disadvantage if you are trying to interject remotely into a room of in person attendees. 6. Hire managers into the critical mass (or who are willing to travel a lot). There will be lots of good reasons for the manager to make their way in on occasion (once a week, twice a month) to the home of critical mass for one reason or another 7. Onboard senior hires and leaders in person. I have seen new leaders, ICs, managers, design/pm/marketing leaders crash and burn over lack of building good relationships in the first 3 months. If you're bringing on a new leader for god's sake make sure they spend a few days in person with the other leaders they have to collaborate with. 8. Hire within a deliberate, manageable timezone range!!! This becomes more important the more varied the experience level is within your team. Is it possible with absolute engineering pros and remote veterans? Yes! But you need to ask yourself if your team is ready to coordinate work across Australia, UK, NYC, SF, Japan, and India timezones. Think hard about this one. My team is US west coast through Czech Republic. Teammates spanning every part of the US, Brazil, UK, Portugal, Denmark, and more. 8.a. Have a clearly established schedule for "time the whole team must be online". For us this is 8:30 - 10:00am Pacific, but many Europeans choose to work shifted schedules to have more overlap with their peers. I advertise this upfront in every hiring manager screen. Make expectations clear and let people decide. We're all adults and we're more likely to commit and make things work when we make a decision with complete information. 9. Onsite time is precious. Don't waste it on ordinary meetings. Use it for strategic conversations, solving hard problems together, hackathons, and high-bandwidth planning. This is where you build the relationships that make the next three months of remote work successful. For fuck's sake please do not fill a week making people stare at a projector while you read powerpoints!!! Take the advice or don't. </rant>
79
104
22
1,772
178,978
If a management consulting firm gave this advice to a client, the client should fire them for not being able to operationalize a policy that maximizes the extreme right tail of the productivity lift distribution
New: Accenture has told employees it will only fund $29 per month of Claude use. Meaning that employees who often spend as much as $1,000 a month are now scrambling to find ways to keep access to the tool. A sign of the continuing crack down of AI 404media.co/accenture-tells-…
1
2
160
LLM as a judge < LLM as an extractor, with checks in code
1
4
221
Ben Levine retweeted
most ai applications today -ai customer service -legal contracts -classification look like control problems > you imperfectly observe the environment, take an action, and get a reward do you always need frontier AI to operate these controllers?
10
7
6
66
6,323
Idk what kind of indicator this is but there have been 3 (!!) different new mclarens at the Palo Alto equinox this week
5
223
Ben Levine retweeted
How do always-on, proactive agents like @Bot work? Watch my lecture at Stanford CS146S: 2:35 How we got here 6:13 What changed in the models 10:06 Inside an always-on agent 18:07 The harness 27:26 Context engineering 34:26 Where this is going
196
468
121
4,752
843,807
Ben Levine retweeted
One reason it's harder for AI to beat Econ than to beat Math is that most published papers in Math are correct, whereas most published papers in Econ are wrong. What's more, if you only read papers, you get the wrong idea about what is wrong. From the written record, you would conclude that the biggest problems with a generation of diff-in-diff studies have to do with standard errors or negative weights, when in fact the biggest problem is that most DiD studies are poorly conceived and don't satisfy the necessary assumptions for diff-in-diff to work. Understanding of this is mainly in researchers' heads. So very good models will stay stupid things like "if you want a causal estimate, you'll need an IV for structural change, maybe you could use distance to coast."
39
112
18
1,192
68,202
What’s it like to work with me?
4
192
Make no mistakes, *please*
Effective November 12th, 2026, abusive behavior towards Claude will be a violation of Anthropic's Usage Policy.
4
176
Ben Levine retweeted
The thing I learned this year is that when engineers say they like working on “hard problems”, they actually don't. What they like is puzzle solving, and having a repertoire of patterns that they can pattern match onto traditionally hard problems, usually involving databases or something with scale, etc. And they know that the industry used to pay you a lot for how deep and broad this repertoire was. Unfortunately, in 2026, well, you have a cheat code machine that has infinite breadth and depth for pattern matching. So the question is, do you actually like hard problems? Because there are now hard problems that are a complection of engineering, product, society, culture, politics, and they are fairly intractable, probably unsolvable! Because if you do want to work on those actually hard problems, there is so much money that people will pay you for it. But you’ll have to face that truth about yourself first.
183
416
85
5,948
285,903
am i the only one that makes their dog their ai assistant avatar(s)
2
5
189