VecGS
Chaotic good
No bio...
User ID: 599
VecGS
Chaotic good
25d ago
That's the thing exactly. If, through using AI tools, you can be the first to prove something, isn't that a good thing? Only through discovery can we reach higher to ask bigger question. It took a lot of work as a civilization to get to the point where we had the language to construct things like the Navier–Stokes equation in the first place; they don't just get written down in a vacuum.
So we've gotten to a step change in the system. The real challenge will be to be the first the ask the question in the first place. Even if that is automated away, you're still left with the pure joy of learning and discovery which is rewarding unto itself.
VecGS
Chaotic good
26d ago
I'm certainly in the "love of the game" category of programmer. I honestly think that you have as many in math. And unlike computers, few people ever went into math to get rich -- so my internal estimate is even higher for the math side.
Side-topic: I think the programmers who are in it for just money are the ones that are going to get selected out. I've been at this for a long enough that when I started a programmer was a good white-collar job, but not one to get you rich. My experience, and frustration, with how the field has changed over the decades is that the quality of people has become absolutely dreadful outside of a few highly selective places. And even places like Amazon (where I spent some time) are compromised.
I get joy from many things. From tweaking things big and small. I don't think I ever really got joy from the act of writing code. It was always a means to an end. And the end doesn't have to be strictly useful. Something I've been kicking around for the past 35 years (also: crap... I'm feeling old) is an implementation of the Game of Life from John Conway who would fit in with your list perfectly. I came up with what I still think is the fastest way to play it with "normal" code in the general case. Using Claude I was able to get this implemented with a front-end that I can interact with. I get joy from the algorithm. Joy from thinking of low-level optimizations. Joy from learning more about the innards of how the processor I'm on handles cache lines.
I don't get joy from setting up a build system. No joy from setting up a front-end that runs on my Mac. Very little joy from tweaking an unrolled inner loop. Less joy from setting up unit tests.
Having AI lets me get to that joy. And it makes me incredibly happy. And it lets me paper over the parts I consider ditch digging.
My gut says that the folks in math want the same thing. Come up with an idea, and have something else handle the proof. There's joy in simply making something better or pushing the frontier a bit more.
And with AI, you can make bigger leaps. Ask bigger questions. Get more joy from discovery.
In case you're interested: https://github.com/gburgyan/vlife and a technical paper in docs that's more presentable.
VecGS
Chaotic good
26d ago
People still play chess even though computers have long been better at it than use humans.
I'm in programming, one of the fields that has already been most affected by AI. I can also report that I'm having some of the most fun and being more productive than I've ever been. My key has been to just embrace AI hard and use it as a tool the best that I can.
You also see the culture war starting up in that space. There are several Linux distros that have people pledge that they haven't used AI in the process of doing their contributions. My prediction is either: a) people are just lying about not using AI or b) they'll just lose and people who are using AI will simply outperform them in the marketplace of ideas.
So math people simply need to get on board. The train is leaving no matter their opinion of it.
VecGS
Chaotic good
1mo ago
Hasn't gotten to threats or anything, thankfully. I'm doing what I need to for another month. It's just so disappointing to see the quality of my "replacement."
I'm giving her remedial training on Windows and how to use AI -- despite the selling point from her firm that they are an AI first shop. I told her to install Copilot CLI (because that's what's authorized to be used, not my choice) and she was doing the equivalent of a blank look of "how?" She hadn't even thought to hit up Google and type "install copilot cli," which of course brings back the correct result at the top.
A few months back I was notified that my employer will no longer need my services in around 90 days from the notice. That 90 days will be up October 19th.
I posted here a few weeks ago about my hardware project which is still going strong! I now have a reproducible process to go from a netlist that describes the connections on a board to doing all of the floorplanning and routing: https://imgur.com/a/3BMLd1I. This is some really dense analog for the lock-in amplifier I was talking about.
Doing this basically more than full time is leading to a bit of burnout. (Being realistic, I'm not doing my day job any more -- it's only training my Indian replacements and I'm not that motivated to do that if you can imagine.)
So, I'm about to release my other side project! A full lifecycle API toolkit: aat! Adaptive API Toolkit.
Presenting: https://gburgyan.github.io/aat/
It handles all sorts of things that get in the way of developing, testing, and integrating with APIs. This addresses the pain points that I've continually run into: local testing of services, testing from a CI/CD perspective, sharing those tests with your colleagues, and even supporting AI-led integrations for people using your APIs.
Here's an example of the usage of this for the travel booking service Duffel: https://github.com/gburgyan/aat-duffel
And it's all open source!
The way I'm figuring, I'm a 53-year-old dude. The AI revolution is coming for my profession: software development. I have no chance to land a job when I'm competing in the traditional way with the tens of thousands of programmers out there. If I do exactly the same thing as everyone else I will be nothing but frustrated and still lack a job. This may be better shot.
Assuming I even want a job of course. The hardware project sounds like it might be more fun anyway.
Thanks! The hardware side has gotten so much (relatively) cheaper and accessible than before... I figure not counting my time (which my present company is paying for now), I'll need to sink $20-30k to have a couple spins and enough stock to hopefully sell. I can cover this myself. I figure this is a bet on myself and I can stack the odds as much as I can.
And you nailed it on the "what's already out there" aspect. My working model is that there are cool chips out there that do neat stuff, the challenge is going from a dev board to something that actually works. One of the more esoteric things I'm thinking about is something like a isolated biopotential interface. It's basically doing the design work to wrap an ADS1299 (https://www.digikey.com/en/products/detail/texas-instruments/ADS1299-4PAGR/6590685) with the appropriate front-end and connect it to a processor. There's no genius in that.
The interesting thing is all from the firmware. To do something useful with that in a non-trivial way -- beyond the typical Arduino/Raspberry Pi toy demos -- everything becomes "ok, we'll need to spin up an RTOS and do a bunch of work" and you give up and buy the instrument that's $1k and move on. The firmware side is cool because it's mostly declarative.
Again, this is nothing really new. LabView had that model back to at least the '90s. The idea of a driver having virtual "wires" being connected is right out of SCADE. But I've not really seen this on a firmware level. The thing that I always run into is even configuring a processor is a pain.
My product is basically:
- A cheap base with the processor -- that can be dynamically configured by the firmware to accept...
- A family of instrument cards that plug in to the three exposed ports.
You then have a declarative profile that can wire up things to do what you need. The goal is to replace a stack of instruments with all of the scripting glue that runs on the host and have it all run on a cheap, trustworthy instrument.
Something kind of like this:
# How much of your 10 uF capacitor is left at 10 V?
#
# Instrument Slot A -- B4 lock-in: a 50 mV, 1 kHz probe, and a phase
# reference to tell the capacitive current from the
# lossy one.
# Instrument Slot B -- A3 SMU: the DC operating point, held and measured,
# with a compliance limit so a shorted part is a
# clamp and not a fire.
#
# Neither card is a better version of the other. The SMU has no synchronous
# detector: it cannot pull a microamp of 1 kHz current out of a 10 V DC
# pedestal. The lock-in has no source with compliance: it cannot hold that
# pedestal. The experiment needs both, on the same tick, and the only thing
# composition costs here is a bias tee in the fixture (100 kOhm from FORCE,
# blocking cap from EXC OUT, the far plate into the TIA's virtual ground).
profile:
name: mlcc_bias_derating
version: 0.1.0
schema_version: 1
tick_hz: 100
ui:
parameters:
- { id: f_hz, type: q16, default: 1000.0 } # test frequency
- { id: v_ac, type: q16, default: 0.05 } # 50 mV -- small-signal, on purpose
surface:
- { wire: var.cap_uf, role: readout, label: "C", unit: "uF" }
- { wire: sig.v_bias, role: readout, label: "Vdc", unit: "V" }
- { wire: smu1.sweep_start, role: action, label: "Run", key: f1 }
variables:
- { id: cap_uf, type: f32, initial: 0.0 }
blocks:
- id: smu1
kind: smu # Slot B
params:
mode: source_v
sense: four_wire_measure # the bias resistor's drop stays out of the reading
compliance_i_a: 0.001
nplc: 10
sweep: { start: 0.0, stop: 10.0, points: 41, spacing: lin, dwell: 2.0 }
out:
v: sig.v_bias # the volts actually across the part
i: sig.i_leak # a second curve, free: leakage vs bias
- id: li1
kind: lockin # Slot A
params:
mode: impedance
ref_source: internal
excitation: sine
input_mode: current_A # the 10^6 V/A TIA takes the cap's AC current
coupling: ac # the 10 V pedestal never reaches the gain chain
sensitivity: auto
time_constant_s: 0.3
slope_db_oct: 12
in:
ref_freq_hz: ui.parameters.f_hz
excitation_v: ui.parameters.v_ac
out:
y: sig.i_quad # autophase first: the capacitive current lands here
theta_deg: sig.loss_angle # and the loss angle comes along for nothing
overload: sig.li1_ovld
tasks:
- id: report
triggered_by: { tick_periodic: 500ms }
body:
- write: { "var.cap_uf": "1.0e6 * sig.i_quad / (6.2831853 * ui.parameters.f_hz * ui.parameters.v_ac)" }
Something like this is obviously "easy" to do if you have both a source-measure unit and a lock-in amplifier and a script. My goal would is to be able to sell something that can do this for maybe $1000, and that leaves a lot of margin.
The other angle I have on this is that something like this is far easier for an AI to write and reason about than normal firmware.
I have an absolute crap-ton left to do. Maybe 60% of the firmware is ready to do this and probably closer to 30% on the hardware side.
VecGS
Chaotic good
2mo ago
Well... a few weeks ago I got laid off. The cool thing is I get 90 days to do "knowledge transfer," then get another five weeks of severance. So, I'm basically set for the year. (Full disclosure: I'm 53. I tried to retire before this, but got sucked back in.) My manager was also laid off and her directive was to basically fuck off and do the bare minimum.
My background is firmly on the software side of the world. Electronics have always been a fascination for me. I've been tinkering with this stuff since the early '80s, but never fully found the motivation to really pursue it for some reason.
So this May rolls around (well before the layoff), and I decide to make a controller board to replace the one in my smoker. It decided to act weird and I got frustrated. This isn't a rational decision because a new one is cheap compared to what I'm doing. I dove in with Claude Code and built a full PCB design platform. (No, I'm not replacing KiCad, et. al. The is the netlist generation side, heavily leaning on SKiDL.) That project grew and grew. Out of control basically. I even laid out a PCB and fabbed it. And it mostly works, amazingly enough. (I also learned to not try to hand-place 0402 parts because that sucks.)
The other side of this is building a complete firmware system from scratch. And now most of that actually works as well. I'm planning on open sourcing this part.
Then, right before I got laid off, I decided to pivot the hardware side. Without getting into too much detail (for fear of abject boredom, not secrecy), the new target is a line of modular, low-cost lab instruments. "Weird" shit. Things that I want to play with. Things like a lock-in amplifier, a source measure unit, precision DAQ (and a bunch of other stuff). That combined with the firmware makes some experiments way easier than the "real" instruments because you have multiple instruments sharing the same core.
Basically, I want to bring down the cost entry for some of this stuff a lot lower. But still have a real calibration story so people can do real science. And to bring the cost of these instruments down in general.
TL;DR: My side project turned into my job (hopefully). Fingers crossed I can pull this off and not lose my shirt too much!
On AI. Without AI this would have been folly to even attempt. My best guess for what this would cost without AI is on the order of $5-10MM to hire a team of experienced devs for a year or two. This is not something I have knocking around in my bank account. If I would have tried to develop this myself (without AI), it would take me the remainder of my life basically. As a bonus I would come off as a crackpot: "I have the perfect way of designing firmware! Just you wait and see!" With AI, I can make this real in a matter of months as an individual.
VecGS
Chaotic good
8mo ago
I've heard of Bad Bunny before, but I've never heard his music before. I'm not a Spanish speaker and I'm not overly big on listening to foreign language music where I'm expected to understand.
The half-time show was... ok. I've seen better. I've seen worse. From what I've read online from native Spanish-speakers, I wasn't missing much and many people are reporting that they couldn't understand it either.
I'll probably never listen to his music again. I didn't find it interesting enough to ever seek out in the future.
(Exceptions to the foreign language music for me include things like German industrial where it's a chill beat for me to program or some classical/operatic music)
I would argue that a large part of art is connecting with another human being through a shared experience. I think that's why so many artists are up in arms about making art so "easy," because it drops the human connection in many ways. The only human in the loop is prompting.
My view is that things will adapt. The human connection is still needed and there will always be art done the "old fashioned way." People still drag around large-format cameras to take pictures or simply shoot on film. My friend does painting and is trying to make a business of it.
At the same time, things have gotten easier. You don't have to be a "good" photographer to take a good photo with your phone's camera -- because there's already so much AI in that loop making things better. Same with so much other human-made art (and beyond). It's so much easier to get to the joy of creation. With photography, you can be creative with your phone and don't need the traditional thousands of dollars of gear to start creating.
The removal of those gates, both financial and the gate of years of practice, is what allows for that creation. I think we should celebrate that, not regulate it. Why does not "paying your dues" make your contribution invalid?
If your art, regardless of how it's made, helps you connect to others, I say that's a good thing.
I have a lot of other ideas in my head I'm trying to get out and make into a coherent argument, but I'm having trouble. For instance, we've had computer spell-checkers for decades now and no one blinks at that. We've had simple grammar checkers for a while as well, predating the AI everywhere thing we're seeing now, and no one really cares about that. It's only a matter of time before AI does more and people don't care. The only thing that really frustrates me is when people give up their agency to AI and stop thinking for themselves. AI is an amplifier of a human's ideas and abilities. But humans are still important.
More options
Context Copy link