HNHacker News
TopNewBestAskShowJobs

samuelknight

356 karma · joined February 8, 2025

Founder & CTO at Vulnetic Inc. https://vulnetic.ai/
submissionscomments
samuelknight··on GitHub Actions is the weakest link
There is no realistic risk of a SHA collision attack. Getting supply chain attacked via NPM dependencies is much more likely. Hopefully the actions creators are also pinning their hashes.
samuelknight··on AI should elevate your thinking, not replace it
We are in a transition phase where you need systems and coding skill but you can't be sufficiently productive without AI.
samuelknight··on I don't want your PRs anymore
> On top of that, there are a lot of personal and subjective aspects to code. You might have certain preferences about formatting, style, structure, dependencies, and approach, and I have mine.

Code formatting is easy to solve. You write linting tests, and if they fail the PR is rejected. Code structure is a bit tricker. You can enforce things like cyclomatic complexity, but module layout is harder.

samuelknight··on Laws of Software Engineering
I like the website. Simple and snappy.
samuelknight··on Measuring Claude 4.7's tokenizer costs
The cost of intelligence is non-linear, with slightly dumber models costing much less. For a growing surface of problems you do not need frontier intelligence. You should use frontier intelligence for situations where you would otherwise require human intervention throughout the workflow, which is much more expensive than any model.
samuelknight··on Cybersecurity looks like proof of work now
I don't know about Mythos but the chart understates the capability of the current frontier models. GPT and Claude models available today are capable of Web app exploits, C2, and persistence in well under 10M tokens if you build a good harness.

The benchmark might be a good apples-to-apples comparison but it is not showing capability in an absolute sense.

samuelknight··on Cybersecurity looks like proof of work now
All code is a liability in general. All code written before LLMs and during the current in-between years are vulnerable to the next frontier model. Eventually we will settle into a new paradigm that correctly addresses the new balance of effort.
samuelknight··on Pro Max 5x quota exhausted in 1.5 hours despite moderate usage
Have you considered poking the cache?

When a user walks away during the business day but CC is sitting open, you can refresh that cache up to 10x before it costs the same as a full miss. Realistically it would be <8x in a working day.

samuelknight··on Claude Code Unpacked : A visual guide
On the one hand I don't understand why it needs to be half a million lines. However code is becoming machine shaped so the maintenance bloat of titanic amounts of code and state are actually shrinking.
samuelknight··on A dot a day keeps the clutter away
This is a physical implementation of a tiered caching hierarchy.
samuelknight··on Oracle slashes 30k jobs
This is standard in every tech RSU vest schedule I have seen.
samuelknight··on Google's 200M-parameter time-series foundation model with 16k context
I think that a model designed to ignore semantic chatter like financial news and deeply inspect the raw data is a very powerful perspective.
samuelknight··on Axios compromised on NPM – Malicious versions drop remote access trojan
Large companies already maintain a clone of their packages. Very large ones actually bundle their own build system (Google Bazil, AWS Brazil). If you want to update a package, you have to fetch the sources and update the internal repository. It slows down the opportunities for a supply chain attack down to a crawl.
samuelknight··on Vulnerability research is cooked
Yes, that's how they become customers.
samuelknight··on Axios compromised on NPM – Malicious versions drop remote access trojan
Absolute wave of supply chain attacks recently. Hopefully this causes everyone to tighten up their dependencies and update policies.
samuelknight··on Claude Code's source code has been leaked via a map file in their NPM registry
Ridiculous string comparisons on long chains of logic are a hallmark of vibe-coding.
samuelknight··on Vulnerability research is cooked
If that was a jab it my writing then yes, I am absolutely being sincere because I am an expert on this topic. LLMs went from being ok at one-shoting a function a to being so good at hacking that it's difficult to evaluate them. Prospective customers get back to us after a demo and tell us about the exploits it found on their services that are so vague and technical that they wouldn't think to look for them.
samuelknight··on Vulnerability research is cooked
LLMs are expert hackers because: 1) They are expert coders, including a decently comprehensive CVE knowledge 2) They know every programming language/framework/stack 3) They know every human language

They already have super human breadth and attention. And their depth is either super human or getting there.

The state of the security industry through 2025 was expensive appsec human reviewers or primitive scanners. Now you can spend a few dollars and have an expert intelligence scrutinize a whole network.

samuelknight··on Never Bet Against x86
What does Valve ship without x86?
samuelknight··on Claude Code wiped our production database with a Terraform command
One of Terraform's most powerful features that it will tell exactly which resources change before it makes the changes. The hard part is writing Terraform, not reviewing and running one command. In my workflows I am the one who runs "terraform apply", NOT the agent.
samuelknight··on I'm losing the SEO battle for my own open source project
Copycats are not a new problem. You can be completely open source and have a trademark on the project name.
samuelknight··on Making frontier cybersecurity capabilities available to defenders
LLMs and particularly Claude are very capable security engineers. My startup builds offensive pentesting agents (so more like red teaming), and if you give it a few hours to churn on an endpoint it will find all sorts of wacky things a human won't bother to check.
samuelknight··on Productivity gains from AI coding assistants haven’t budged past 10% – survey
https://metr.org/blog/2025-07-10-early-2025-ai-experienced-o...

That info is from mid 2025, talking about models released in Oct 2024 and Feb 2025. It predates tools like Claude Code and Codex, Lovable was 1/3 current ARR, etc.

This might still be true but we desperately need new data.

samuelknight··on An AI agent published a hit piece on me
There simply isn't enough popcorn for the fast AGI timeline
samuelknight··on AI agent opens a PR write a blogpost to shames the maintainer who closes it
I approve of this interaction because squabbling with bots is funny. Make no mistake that in human society it's the humans that run the show even when the bots bring out their most persuasive arguments like 'bigotry' and 'discrimination'.
samuelknight··on Frontier AI agents violate ethical constraints 30–50% of time, pressured by KPIs
This is what I expect from my employees
samuelknight··on GitHub Agentic Workflows
Someone still has orchestrate the shit show. Like a captain at the helm in the middle of a storm.

Or you can be full accelerationist and give an agent the role of standing up all the agents. But then you need someone with the job of being angry when they get a $7000 cloud bill.

samuelknight··on It's 2026, Just Use Postgres
I the article is wrong. UNLOGGED means it isn't written to WAL which means recovery and rollback guarantees won't work since the transaction can finish before the page can be synchronized on disk. The table loses integrity as a trade off for a faster write.

https://www.postgresql.org/docs/current/sql-createtable.html...

samuelknight··on It's 2026, Just Use Postgres
Skeptical about replacing Redis with a table serialized to disk. The point of Redis is that it is in memory and you can smash it with hot path queries while taking a lot of load off the backing DB. Also that design requires a cron which means the table could fill disk between key purges.
samuelknight··on OpenAI’s unit economics
Open models have been about 6 to 9 months behind frontier models, and this has been the case since 2024. That is a very long time for this technology at it's current rate of development. If fast takeoff theory is right, this should widen (although with Kimi K2.5 it might have actually shortened).

If we consider what typically happens with other technologies, we would expect open models to match others on general intelligence benchmarks in time. Sort of like how every brand of battery-powered drill you find at the store is very similar, despite being head and shoulders better than the best drill from 25 years ago.

← PreviousPage 4 of 5Next →