HNHacker News
TopNewBestAskShowJobs

theChris-in

5 karma · joined May 20, 2026

I’ve spent two decades working on VoIP, real-time systems, and backend architecture, building systems that scale and exploring the trade-offs behind them. I care about clarity in code, restraint in design, and deep technology that doesn’t announce itself.

https://thechris.in

submissionscomments
theChris-in··on Banned Words List
I think our priorities are so effed right now smh
theChris-in··on What 50 open source projects taught us about security in the AI era
I have a genuine question. Why can't any of these analyses on security ever throw up a gradual adoption of formal methods? Like, wouldn't that be the next thing that people should slowly start getting used to?
theChris-in··on Is the industry ready for tokens-constrained work?
I think eventually we will converge on a new AI + old AI (ILP, ASP) way of working.
theChris-in··on Banned Words List
I don't understand. Is "woman" not allowed to appear in any govt. website? What??
theChris-in··on Ask HN: Anyone interested in building a harness-only benchmark?
> should test the harness capability rather than model's knowledge/capability.

Then we need a provenance for model inference, generalized. This should be interesting. We would be trying to deterministically generalize a baseline "can do this" for models... Maybe categorize by parameter class.

theChris-in··on Ask HN: Anyone interested in building a harness-only benchmark?
We can do a mix of general use (as in user stories) plus a few academic benchmarks.

So you have any specific ideas?

You can hmu at iam@thechris.in

theChris-in··on Ask HN: Anyone interested in building a harness-only benchmark?
Interesting idea. If you can manage the infra, I can put together a replicable test suite.