Floppy disks were ubiquitous when I was in college. When I got into Linux, I did an experiment raw writing zeros to floppies with dd to see what percentage of them had I/O errors. I tested with a stack of about 50 of them that were left in our computer lab over the years (different brands). The failure rate was staggering. Something like 30-40% of them had bad sectors. After that, I realized that I could never rely on them as a storage medium for anything important without regular backups.
We could probably debate this ad nauseam, so I'll just give you my most compelling arguments.
1) Writing code the "old fashioned" way (i.e., a Python program that does X, Y, Z) allows you to arrive at a battle tested solution that will not change over time. From a risk assessment perspective, the behavior is essentially immutable, allowing a business to guarantee consistent behavior over long periods of time.
2) Just because something hasn't happened to you do, does not mean that it will not happen. LLM are opaque. If you stay on the "happy path", you may see consistent behavior for long periods of time, but there's always potential for an edge case where something goes catastrophically wrong. This is without even opening the can of worms regarding prompt injection and intentional sabotage of a working system.
3) There are plenty of real world examples of an LLM spontaneously deleting data from a DB (or the entire DB) or otherwise going completely off the rails. These might seem hyperbolic, but it happened at our company (to a test DB, not production). The severity of errors that occur can be existential to a business' survival without the proper guard rails.
4) There's no concrete way to truly confirm understanding between an LLM and a human. It can tell you that it completely understands what you want, and then it can do exactly the opposite. Followed by, "my bad" (Claude's new favorite catch phrase). Code can be audited and even proven to be correct given the appropriate level of time and energy.
My best results have been gleaned in using LLM to produce deterministic systems. I recognize everyone has different use cases and needs, but this seems to be the best use of the technology in my experience.
The craziest for me is companies that sticking stochastic agents into automated business processes and expecting stable/reliable outcomes. Businesses want deterministic processes in the vast majority of cases.
I have been working on a language learning app for myself, and I am using a textbook that I like as the basis for an Anki inspired “learning tree”. This is working pretty well because I can build progressions from the original table of contents.
A huge chunk of SoaceX value in their filing is attributed to their AI technology (aka Grok). I believe it’s 90% or more… Now, it seems they’re leasing the infrastructure required for Grok to scale to Anthropic and Google. I wonder how that math works…
Neurotone.ai || Remote (Americas East / Western Europe) || Full-time || neurotone.com
We build AI-assisted auditory training for people with hearing loss. Our product is live, in clinics, and growing internationally.
Backend Engineer: Postgres-heavy work on Supabase: PL/pgSQL functions, pgmq queues, Deno edge functions, billing and cron systems. You like databases enough to enjoy writing triggers and RLS policies, and you want to own systems end-to-end rather than just ship tickets.
QA Specialist: Manual testing of our iOS/Android app and web tools, working alongside our head of QA on a clinical product where data quality, platform stability, and accessibility are critical to our mission.
Note: Candidates must be in time zones no further west than EST; Western Europe is fine.
If interested, contact travis AT neurotone.com. Join us in making a difference!
I tend to agree, but the gains will come at the expense of the early adopters. Then again, this has been the case in so many industries throughout history.
Overkill in what way exactly? The LOC of the project shouldn't have any bearing on most people's usage of the project. SQLite is one of the well tested and mature projects in the world. What exactly would motivate someone to use PeakSlab instead? What problem are you solving?
https://parcero.ai/
https://verbose.ly