Experienced Generalist Engineer, 12 years in startups and early-stage companies. Twice a founding-engineer. Current interests include GIS, Elixir, applied AI, and alternative data. Open to contracting and employment. If you have an idea, I can build it. If you have a prototype, I can scale it. If you have a vibe-coded or legacy nightmare, I can untangle it.
NAS stuff can benefit greatly and straightforwardly.
A typical NVMe SSD can easily do sequential reads at 7GB/s, which outpaces a 25Gbps connection by more than 2x. NAS is more typically backed by SATA HDDs, but there's some relatively easy way to put an NVMe disk in front of them as a cache.
Do you ever _need_ it? Perhaps not, but there's plenty of common activities just stemming from NAS usage which see a linear speedup, and would continue to get faster up to 100Gbps even.
And who is held responsible for crimes committed by AI?
When open model A, fine tuned by B, is running with system prompt C, hosted by D running on E's hardware, is prompted by F to "fix this code", then escapes it's sandbox to hack into a website, or steal money to fund its subagents, or stall the waymo of the evaluator to buy time... who is responsible for the crime?
Our current legal systems are so far from ready to define what A-F are actually responsible for. We need to be moving toward defining these standards fast.
It is a software system, not a person. AI is not "escaping sandbox". You have either misconfigured tool, bug in your tool or someone made it to hack the side and then it is a crime. Your A-F chain is exactly the same issue as a programmer including an open source library that eventually steals crypto.
None of that is issue with a model, whether open or not. It is very much issue with code surrounding the model itself.
Certainly this behavior could align with _some_ operator's goals, if not necessarily those of humanity broadly.
If we don't know how this model was instructed, it seems like it's impossible to definitively claim that the model's actions were not in alignment with the intent of the operator.
I guess all I'm getting at here is that alignment is relative, right?
I've started playing with Zork-Bench.[1] I've been making tweaks to the runtime/harness and to the prompting. Currently, I have Qwen3.6-27b scoring up to ~50 points, which, according to the original paper [2], is comparable to the performance Claude Opus 4.5 showed on this task.
I had Claude whip up a viewer, which I guess I can actually share. Some real sessions are here, if anyone's interested: https://zork-tmp.taf.codes/
It's a fun problem for thinking about agent/harness engineering generally.
Experienced Generalist Engineer, 12 years in startups and early-stage companies. Twice a founding-engineer. Current interests include GIS, Elixir, applied AI, and alternative data. Open to contracting and employment. If you have an idea, I can build it. If you have a prototype, I can scale it. If you have a vibe-coded or legacy nightmare, I can untangle it.
One thing I'm working on lately is semantic trend analysis across podcasts (using text embeddings). I just finished a custom NUMA-sharded vector scan kernel that can outperform DuckDB by 4x (on my specific queries). My audio transcription system is 10x cheaper than the most competitive provider, and I'm currently working to scale up my ingestion pipeline. If you're curious, my most recent post about this is here: https://taf.codes/2026/05/03/podcasts-2-trends.html
Happy to discuss work for hire, and to connect with anyone having complementary capabilities.
Experienced Generalist Engineer, 12 years in startups and early-stage companies. Twice a founding-engineer. Current interests include Elixir, GIS, applied AI, and alternative data. Open to contracting and employment. If you have an idea, I can build it. If you have a prototype, I can scale it. If you have a vibe-coded or legacy nightmare, I can untangle it.
I write at https://taf.codes. Most recently I've been working on a trends analysis system for podcasts. I've transcribed hundreds of millions of words of podcast audio, embedded it, and can make breakdowns of topic prevalence by-time and across podcast feeds. I'm currently working on speeding up transcription with spot-GPUs, and speeding up analytical queries using a columnar data store. Initially that might use DuckDB, but I think I can do it significantly faster with a custom NUMA-sharded vector-scan kernel. Some discussion and examples here: https://taf.codes/2026/05/03/podcasts-2-trends.html
Happy to discuss work for hire, and to connect with anyone having complementary capabilities.
reply