datameta

Ah so lexically close to Aleksii in Ukrainian or Alexei in Russian. I plumbed my knowledge but it may have sounded absolutist.

However, colloquially, I think most people are not aware of Balkan naming conventions. How do you pronounce Aleksa in Croatian?

Aleksandr (diminuitive being Alyosha) is male. Aleksandra (Sasha) is female. I've never heard anyone male or female called "Aleksa". Perhaps in some regional dialect but doubtful. There's Alik (pronounced AH-l'eek) as a shortened form for some middle-aged men in russian-speaking central asia, and Oleg in russia (pronounced ah-L'EHG).

In the russian diaspora in the US, Alex is pronounced AH-leks. If there was an analogous Alexa (there isn't) the pronunciation would be ah-LEK-sa like the service.

I wonder how many companies had this play out? IBM is another famous example of internal rot of engineering excellence in the 2000s as the business shell was gives fresh coats of paint and varnish.

What of the "just so" attitude of a child growing up? Everything is taken at face value, there is no comparison, only stories (unless you have a catalogue of 30EB of 8K earth footage or something to that effect for them to fawn over). They don't have the reference frame for other situations for a while, perhaps long enough to not be able to see things differently?

This makes me think of multi-generational migrations north out of Africa. There's only so much that can be passed orally losslessly. Eventually the group in north siberia after 20K years doesn't see living any other way.

Aspartame is linked to anxiety.

I did some cursory snooping and it looks like it could be that both units should be MV/m

I still experience what you outline for bootstrapping understanding, or bringing the knowledge horizon closer, so to speak. Part of the learning though is through the vigilance of parsing and correcting.

When I had a small passive circuit in my head that I wanted to one-shot solder on a protoboard and didn't want to get bogged down in KiCAD, talking through it with an AI and repeatedly correcting its understanding really solidified my own. It's like mentoring while being tutored.

So I still see value in use for smaller novel projects, but using LLMs to hit a deadline for production is not something I want to do any longer, for the time being.

The more I use LLMs to code, the farther I stray from the joy of coding.

I won't regurgitate the common sentiment of it being like pair programming with an oddly spiky profile junior engineer savant. I find my emotions being activated too often in having to recontextualize or reign in or re-orient or repeat.

If I wrote code that doesn't work, that is on me. If I have to read code that doesn't work, I didn't write it, and it's veracity or accuracy are only uncoverable by me - but are already being treated as gospel by my AI peer, and the LLM keeps asserting issues with "my" code that I don't yet feel I have ownership over because I have not ingested it fully is supremely tiring. The whiplash feels almost like an abusive relationship.

I want to get back to feeling accomplished once finished with a coding session, not like I just went through a kafkaesque ringer.

I've found that having tasks last 10 minutes is an activation energy economizer (with option to extend 2x10 or 3x10 if in flow). It also simplifies the overhead of deciding how long something should take. I think the key takeaway from it all is that one would benefit most from iterating on their system of choice, keeping what works and carrying it forward, instead of doing impassioned rip-ups or switching to entirely new systems frequently

This makes me think of a reduction gear as a metaphor. At a high enough ratio, the torque is enormous but being put toward barely perceptible movement. There is the huge amount of computation happening to result in SVG that resembles a pelican on a bicycle.

I agree with the last two sentences but simultaneously think that starting to defacto believe you cannot have an equal or better solution compared to the AI is the start of atrophy of those skills.

This seems like a good takeaway to me

I could be mistaken with this nitpick but isn't there a unit mismatch in "...just 20 watts—the same amount of electricity that powers two LED lightbulbs for 24 hours..."?

SEEKING WORK | USA REMOTE | NYC HYBRID

  Technologies -
  Programming: C/C++, Python, JavaScript, Kotlin, VHDL, SystemVerilog
  Embedded Systems: PowerPC, ESP32, STM32, iCE40, ECP5
  ML: TensorFlow Lite Micro, PyTorch, OpenCV, MATLAB, ONNX
  IoT: BLE, WiFi, MQTT, LoRa

  Resume: https://docs.google.com/document/d/18mjB1tIH5s-ZmiET_smUxbMo3FaiC637aW_us8P0U5I
  Email: dm@datameta.info

Hi, I'm an embedded systems engineer/architect bridging the full compute spectrum of ultra-low power IoT, edge systems, and HPC.

My latest projects have been in full-stack edge ML R&D, memory controller firmware dev for POWER servers, and chip bringup/testing for Z mainframes.

I'm passionate about modular synthesis and tinkering with all sorts of analog circuits and DSP. Currently upskilling in analog PCB design and refining my abilities in FPGA development. Also WIP on a 3D-printed robot arm to practice control systems and actuation.

Build your own (small, diy) robot. Practical applications are the primary method of learning imo.

SEEKING WORK | USA REMOTE | NYC HYBRID

  Technologies -
  Programming: C/C++, Python, JavaScript, Kotlin, VHDL, SystemVerilog
  Embedded Systems: PowerPC, ESP32, STM32, iCE40, ECP5
  ML: TensorFlow Lite Micro, PyTorch, OpenCV, MATLAB, ONNX
  IoT: BLE, WiFi, MQTT, LoRa

  Resume: https://docs.google.com/document/d/18mjB1tIH5s-ZmiET_smUxbMo3FaiC637aW_us8P0U5I/edit?usp=drivesdk
  Email: dm@datameta.info

Hi, I'm an embedded systems engineer/architect bridging the full compute spectrum of ultra-low power IoT, edge systems, and HPC.

My latest projects have been in full-stack edge ML R&D, memory controller firmware dev for POWER servers, and chip bringup/testing for Z mainframes.

I'm passionate about modular synthesis and tinkering with all sorts of analog circuits and DSP. Currently upskilling in analog PCB design and refining my abilities in FPGA development. Also WIP on a 3D-printed robot arm to practice control systems and actuation.

What if one is okay with walking away and letting it run (crawl)? Then context can be 128 or 256GB system RAM while the model is jammed into what precious VRAM there is?

Luckily outside of mission critical systems, like in demoscene coding, I can happily use "44/7" as a 2pi approximation (my beloved)

Can system RAM be used for context (albeit at lower parsing speeds)?

Oh, of course, the timestamp could instead be metadata!

I like to bring up a similar anecdote to the faux pas that just happened. For example, a friend taking a wrong turn, adding time to the ride and making the party somewhat late. It halted the conversation in the car, and to ease the tension I pointed out how I "do that at least once per every drive to X" to try to make the driver less self-conscious and so the other passengers aren't sitting in awkward silence.

I think Buddha would feel it, sit with it, and let it go on its own.

This might be completely naive but can a reversible time component be incorporated into distinguishing two hash calculations? Meaning when unpacked/extrapolated it is a unique signifier but when decomposed it folds back into the standard calculation - is this feasible?

I think what you're describing is well encapsulated by a term I heard a few days ago - boreout.

I'll never forget the UNIVAC's mercury delay line memory in the Computer History museum - that just absolutely blew my mind and showed me how non-trivial of a problem memory was in early computers before even core came around

The magnetic drum memory/storage is a fascinating precursor to RAM and hard drives, performing both functions simultaneously. The vacuum tubes only store the next set of processor instructions to be executed

Off the top of my head: Kolmogorov complexity, Shannon entropy? I suppose also it depends on how faithfully or to what granularity one has extracted information from a natural phenomena

See: Маршрутка (Marshrutka), Colectivo, Matatu

I think sometimes because art needs no why.

SEEKING WORK | USA | REMOTE

  Technologies -
  Programming: C/C++, Python, JavaScript, Kotlin
  Embedded Systems: PowerPC, ESP32, STM32, Altera MAX10, Lattice iCE40UP5k
  ML: TensorFlow Lite, PyTorch, OpenCV, MATLAB, ONNX
  IoT: BLE, WiFi, MQTT, LoRa
  Digital Logic: VHDL, Verilog

  Resume: https://docs.google.com/document/d/18mjB1tIH5s-ZmiET_smUxbMo3FaiC637aW_us8P0U5I/edit?usp=drivesdk
  Email: dm@datameta.info

Hi, I'm an embedded systems engineer/architect with multidisciplinary expertise in software, hardware, IoT, and machine learning. Experienced im low-power remote applications, mid-range edge compute, and HPC.

My latest projects have been in full-stack edge AI R&D, memory controller firmware dev for servers, and chip bringup/testing for mainframes.

I'm passionate about modular synthesis and tinkering with all sorts of circuits and DSP. Currently upskilling in analog PCB design and refining my abilities in FPGA development. Also WIP on a 3D-printed robot arm to practice control systems and actuation.