I've wondered this. We have chain-of-thought, harnesses, etc. — workarounds of a sort due to lack of core model capabilities. But I am very curious if much better next token prediction would simply obsolete that whole setup or not. Either way, the answer would be very revealing.
I don't know about having this on as the default, but it definitely resolves a frequent annoyance of mine. I'd prefer Claude to keep going as far as possible until it's completely blocked. That said, I only give instructions to do that in environments where there's a clear upper bound on "maximal damage" that could be incurred by doing the wrong thing. In live production systems, you really don't want Claude doing much other than observing and reporting anyway.
I've been using a SQLite DB to organize actionability so I don't get stuck in the way that GitHub issue tries to work around. Any questions about which route to take that arise during agentic work are logged to a queue along with a set of plausible candidate routes, a probability assigned to each candidate of whether I will choose that option (including "other/none"), a "resource cost" assessment of the route (e.g., token spend, time), a "stability cost" (e.g., high potential to disrupt things or mainly self-contained), and a set of tasks that describe any downstream work that is dependent on the route chosen.
What Claude does next then depends on the results of a tiny optimization program that tries to maximize the expectation value of agent productivity per unit resource (tokens, time, etc.) conditional on how long it will take me to answer the question (e.g., if Claude has a question for me at 1 AM, there probably won't be a response for another 6-7 hours).
"Agent productivity" is of course a bit nebulous, consisting of a somewhat ad-hoc amalgamation of factors, but in general Claude's actionability loosely corresponds to cases like:
- 2-3 possible routes, each with roughly equal probability of being the one I select, low resource costs, minimal risk of instability, few downstream dependencies: implement each route in parallel
- 2-3 possible routes, one with a much higher probability of being chosen, minimal risk of instability, many downstream dependencies: implement just the top route
- Hundreds of possible routes: block until user response
- 1 possible route, high risk of instability, many downstream dependencies: block until user response
Generally speaking, there should be an active queue at all times and agents should be working on anything that's not blocked in the queue with maximal parallelization.
I’ve recently had a massive productivity boost in my Claude workflow simply by asking it to search for and review: relevant PRs via `gh` search, relevant Slack threads, relevant Confluence pages (+edit history), relevant Jira tickets, relevant Sharepoint docs, and relevant Teams transcripts. I ask it to do this comprehensively before the task, during the task, and after the task. It feels like a superpower.
Haha, I just started doing this on my own. Found it helps the agents preserve state better. I typically ask them to design a DAG first based on a set of specifications and then execute it (each step stores something in a SQLite DB). Iteration is pretty simple then because I just ask for a tweak to one or two steps of the DAG, and then to re-run.
Funny how people are independently converging on similar patterns of "what works" here. Still feels like we're in the wild west with all these ad-hoc patterns of agent orchestration that people are coming up with.
I think it depends on the context and also not being deceptive. If someone just spent 4 hours trying to root cause a SEV with Claude and they finally have a nice high-level Claude-generated summary of all that work, just paste it and share it. Don't waste time trying to reword it to make it seem like you wrote it. A simple "After spending a few hours with Claude, here's the conclusion about what the problem was: [paste]".
On the other hand, if you send someone a very personal and heartfelt message and receive a reply like "Yeah, it was so nice spending time with [niece] today!", well, that's a bit different...
I think this question of what sentences can have truth value attached to them is significant. The liar paradox (a sentence in a formal language stating itself to be false) clearly doesn’t have a truth value. A statement that a specific program halts, however, does seem to be either true or false, regardless of whether or not a general algorithm exists that can answer such questions. In a sense, all sentences that fall on the arithmetical hierarchy seem to me to intuitively have a Boolean truth value (in the standard model, which we assume corresponds to what a program would actually “do” if we ran it forever).
Set theoretic questions like AC or CH are much more difficult for me to intuitively grok in the same way, because they don’t seem to “obviously” be either true or false. You can take either and still end up with a (presumably) consistent theory.
> Consciousness the the fundamental reality; it is the only thing we know for sure.
> I know for sure what I am perceiving
This reflects my view. And I’ve always found it mildly amusing that beings I cannot prove to myself are perceiving attempt to convince me that I’m not perceiving, when that’s exactly what I’m maximally sure of. Imagine arguing with an LLM designed to convince you that you’re not real. It would be weird, wouldn’t it?
I feel like we still don’t have great research on how much of this is due to biological factors related to age and how much is due to confounding factors.
I suspect that sustained creativity may be a result of continuous exposure to new experiences and concepts (which younger people are naturally situated to encounter quite often), so I try to consistently add novelty to my life as I get older, specifically targeting things way outside my comfort zone or previous interests.
For example, I’ve always been sort of uneasy with flying, so I figured I would sign up for general aviation classes and learn to fly myself, which is something I never would have had the slightest inclination toward when I was younger. I ultimately didn’t go through with it, as while signing up, my wife strongly insisted that I find a different form of “novelty” to pursue, but I think it decently illustrates what I was attempting to accomplish.
Some more mundane examples include listening to music that I don’t enjoy, completely mixing up how I dress after decades of wearing the same thing, reading books opposite my interests, taking classes in fields different than what my degree was in, and trying to constantly meet people who are very different from myself.
Weird thought: someone born in the 1800s was (most likely) alive when the first transformer model ran.
Emma Morano died April 15, 2017, the NIPS submission deadline for "Attention Is All You Need" was May 19, and a Wired article indicates they were testing models for quite a few weeks before then.
I wonder how much of the programming language problem is due to churn of the user base. Looking over many comments in this thread, I see “Oh, back when I did HPC...” I used Titan for my own work back in 2012. But after my PhD, I never touched HPC again. So the people writing the code use what’s there but don’t stay long enough to help incentivize new or better languages. Now on the hardware side (e.g., design of interconnects), that more commonly seems to be a full career.
The other issue is that to really get the value out of these machines, you sort of have to tailor your code to the machine itself to some degree. The DOE likes to fund projects that really show off the unique capabilities of supercomputers, and if your project could in principle be done on the cloud or a university cluster, it’s likely to be rejected at the proposal stage. So it’s sort of “all or nothing” in the sense that many codebases for HPC are one-off or even have machine-specific adaptations (e.g., see LAMMPS). No new general purpose language would really make this easier.
Is there not some concept that utilizes cryptography in a way such that information about people is accessible, but if it's accessed, then the access request is added to a ledger (akin to blockchain) such that who made the access, when, and about whom becomes provably public knowledge?
> they all have an obvious and immediate majesty to them.
"Grandeur" is not the only criteria for nice national parks. I'm from the east coast, and while all of the breathtaking views in California were amazing, after a few years of living there I began to get frustrated that I couldn't find anywhere "cozy" to visit during the weekends. Some locations along the Russian River probably came the closest, but the jagged rocks and coniferous trees still didn't manifest the sort of "warm and snug" feeling one gets while river tubing along a mountain river in the Blue Ridge mountains. Temperature deciduous rainforests are actually quite rare across the planet, and particularly when the leaves change colors, it's a sight to behold.
Would be interesting to think about what works are currently out there, published, yet will not be recognized as great intellectual achievements until much later after the fact for some reason.
I’ve heard that bitterness affects children more intensely. So I wonder how much of it is an acquired taste vs bitterness just becoming “milder” over time.
It depends what I'm working on. If it's a bunch of interdependent systems that involve a large amount of data, a giant monitor is better. If the giant monitor is being used to make visible more application surfaces (Slack, email, VS Code, etc.), it makes focus worse.
The biggest improvement I've found for my focus is to force myself to close any open tabs/windows that are not absolutely necessary roughly every two hours. I used to be one of those people with 800 tabs open in the browser and 20 application windows spread across 8 desktop spaces. Was a concentration mess. Requiring myself to "clean up" periodically has really helped.
I set up my own home network with a Vertiv Liebert Li-ion UPS a few years ago and was thinking about how inefficient the whole process is regarding power. The current goes from AC to DC back to AC back to DC. Straight from the UPS as DC would work much better, and as I was teaching myself more about networking equipment, I was surprised to learn that most of it isn't DC input by default (i.e., each piece of equipment tends to come with built-in AC-DC conversion).
Then I started routing ethernet with PoE throughout my house and observed that other than a few large appliances, the majority of powered devices in a typical home in 2026 could be supplied via PoE DC current as well! Lighting, laptops, small/medium televisions. The current PoE spec allows up to 100 W, which covers like 80% of the powered devices in most homes. I think it would make more sense to have fewer AC outlets around the modern house and many more terminals for PoE instead (maybe with a more robust connector than RJ45). I wonder what sort of energy efficiency improvements this would yield. No more power bricks all over the place either.
I’ve always wanted kids, ever since I was a kid myself, but I was never really sure what it would be like to be a parent.
Turns out it’s quite strange, because my kids bring me more joy than anything else. I’ll sit there for hours watching them play. You may think “that’s not strange—tons of parents say that”, but for my sort of personality, it’s very strange. I’ve always thought of myself as sort of overly analytical, detached, ambitious, and a bit obsessive. Not the sort of touchy-feely person who chases a two year old around with a smile on my face and likes watching videos of cute babies. Yet here I am. I enjoy it so much I’ve even tried to figure out if there’s a way I can take a sabbatical from work to spend the last two years with my youngest at home before he goes off to school (seems unlikely given how questions about a random two year gap on my resume might affect my long-term career).
It’s funny that as a kid I always wanted to work at a tech company for the interesting tech, but now as an adult my favorite thing about it has been the 4 months of parental leave I was able to have with each newborn.