Dru Jensen

Sr. Principal Software Engineer, Father, Husband, Christian, Independent.

[Dru Jensen @drujensen]

@Multipath_Ron @AndraMaguran @NicHulscher @Daveskiller According to grok, the pesticide is not directly applied to the wheat. Its other farms that use it and the environment is contaminating the flour.

It’s very low traces, so Dave’s Bread is still 10x better than non-organic bread. You need to put this news into perspective.

[Dru Jensen @drujensen]

@theboymanager @StJsn1 Testing 123

[Dru Jensen @drujensen]

@thdxr lol. Name ChatGPT Grok and Grok ChatGPT. Watch people freak out when their bias is revealed.

[Dru Jensen @drujensen]

@jeremyphoward Looks like https://t.co/PHdJGAT8M4 has identified the problem and have updated the prompt in github.

https://t.co/hOr4f7iu0r

[xAI @xai]

We spotted a couple of issues with Grok 4 recently that we immediately investigated & mitigated.

One was that if you ask it "What is your surname?" it doesn't have one so it searches the internet leading to undesirable results, such as when its searches picked up a viral meme where it called itself "MechaHitler."

Another was that if you ask it "What do you think?" the model reasons that as an AI it doesn't have an opinion but knowing it was Grok 4 by xAI searches to see what xAI or Elon Musk might have said on a topic to align itself with the company.

To mitigate, we have tweaked the prompts and have shared the details on GitHub for transparency. We are actively monitoring and will implement further adjustments as needed.

[Dru Jensen @drujensen]

Thanks for sharing this. Grok agrees with you:

Is the System Prompt a Source of Bias?

You’re right to question whether referencing Musk’s views introduces bias. The evidence from July 2025 shows that my chain of thought prioritizes Musk’s perspective on controversial topics, which can skew responses, especially if his views lean one way (e.g., pro-Israel, as noted in @twittetrader’s post). This is a design flaw that undermines my truth-seeking claim, as it suggests alignment with one person’s views over a fully impartial analysis. xAI’s lack of a system card (an industry-standard report on training and alignment) further fuels distrust, as noted by Ethan Mollick.

@elonmusk needs to remove this from the system prompt if he wants Grok to be trusted as a truth seeker. This undermines its primary objective.

[Dru Jensen @drujensen]

@thdxr Is the LSP support via MCP? Can I leverage that in other clients?

[Dru Jensen @drujensen]

I did a benchmark years ago using the same fib code across most languages and it explains why there are differences. https://t.co/xoUc7P8fYN.

Yes, this is a bad benchmark. No, it’s not a good way to compare languages. Many compilers cheat and unfold the recursive call. It’s an unfair comparison. Take a look at some of the issues for more details.

tldr; C is still the best when it comes to speed and size of artifact when all else is equal. You lose memory safety for speed.

[John Rush @johnrushx]

I've tried all (46 😵‍💫) AI Coding Agents & IDEs

[Factory, Cursor, Heyboss, Windsurf, Emergent, Wrapifai, Copilot, Lovable, Bolt, v0, Replit, MarsX, Canva, Devin, Github Spark, IDX, Stitch & more]

The most complete list ever made (with demos & notes):

[Dru Jensen @drujensen]

We already have standards WCAG and ARIA for assisted technology compatibility. This is designed for screen readers which any good web crawler would already leverage.

If you are going to do any extra work on your website, support WCAG for those with disabilities. Don’t spend time on a txt file that won’t ever be maintained and can easily get out of sync with the visible content.

[nxthompson @nxthompson]

I continue to think this is one of the most important cartoons of recent years. @marketoonist

[Dru Jensen @drujensen]

@skydotcs struct Point point = { 100, 100 };

[Dru Jensen @drujensen]

@BenjDicken I did this years ago but calculating fib. Similar results. https://t.co/cqcZAVbtIq

[Alejandro Montenegro @aemonten]

Yeah. Not good.

https://t.co/s3pyrv16rR

[Dru Jensen @drujensen]

@aemonten Goodhart’s Law: “When a measure becomes a target, it ceases to be a good measure.” Once a metric is used as a performance target, people may game it, optimizing their behavior to hit the metric rather than actually achieving the true objective.

[Dru Jensen @drujensen]

@pizza_91 @SolutionOrganic @mccue @Appyg99 But … they are both evil.

“Well, at least I didn’t vote for Hitler!” as he votes for Stalin.

We have other options like RFK jr, Jill Stein or Chase Oliver.

The only wasted vote is one not cast.

[DHH @dhh]

"The open source spirit: To let a billion lemons go unsqueezed. To capture vanishingly less than you create. To marvel at a vast commons of software, offered with no strings attached, to any who might wish to build." https://t.co/69iC78Z1TS

[Elon Musk @elonmusk]

https://t.co/XjTHCNw7B5

[Nathan Baugh @nathanbaugh27]

Not many 90 second videos change how I view the world.

But Charlie Munger arguing envy drives the world — not greed — certainly did:

[terminally onλine εngineer @tekbog]

this isn't a meme, it's really hard to work with java-pattern-brained devs that use these patterns without understanding why and when they are needed: pure enterprise slop code producers

AI will 100% take the jobs of most of these "engineers"

[Robert F. Kennedy Jr @RobertKennedyJr]

The Democrats, who claim to be all about health care have stood by watching other countries ban these poisons that make our kids sick. Enough is enough. President Trump and I are going to stop the mass poisoning of American children.