Rohit Yadav

Building AI-powered data systems.

[Rohit Yadav @rohitdev_]

Speed of writing code is no longer a bottleneck.

[Rohit Yadav @rohitdev_]

Lost an afternoon to what looked like an off-by-one.

The loop was right. The input wasn't.

np.arange(1, 1.3, 0.1) returns 4 elements, not 3. Float accumulation drifts the last value past stop.

For evenly spaced floats, use np.linspace: specify the count, not the step.

[Rohit Yadav @rohitdev_]

Stop making users do mental gymnastics in your data pipelines.

I just renamed `is_not_historical_data` to `is_nrt`.

Every read used to force a double-negation.

Now the queries actually read like English.

[Rohit Yadav @rohitdev_]

Everything you say will be judged by an LLM.

[Rohit Yadav @rohitdev_]

https://t.co/Sfx3I86BPZ

[Rohit Yadav @rohitdev_]

Which Agent Harness do you suggest using with Grok?

[Brian Roemmele @BrianRoemmele]

Anyone telling you that you need to use Claude to run just about any OpenClaw spending a wacky $1000 or more per month is no expert you should trust.

Use @Grok. Great pricing and 4.3 is in another category. You won’t be burning $1000 a month.

You will absolutely thank me.

[Rohit Yadav @rohitdev_]

Design is an act of creativity and LLM tools are not creative yet.

[Rohit Yadav @rohitdev_]

If you can think, you can build.

[Rohit Yadav @rohitdev_]

A new idea, if not moved forward quickly, wilts.

[Rohit Yadav @rohitdev_]

You’ll really start to notice these limitations when you’re building an AI Agent (interacting directly with a model API) or when you're running a local model. In those cases, you’re dealing with a pure text predictor, and that’s where the "math logic" can easily break down.

[Rohit Yadav @rohitdev_]

Tip for doing Mathematical Calculations:

Instead of asking an LLM to perform calculations directly on a dataset, ask it to write the Python code to do it for you. You can then run that code to get the actual results.

[Rohit Yadav @rohitdev_]

While apps like ChatGPT and Gemini now have built-in tools for searching the web or doing math, it’s easy to forget that the underlying model is still just predicting text.

[Rohit Yadav @rohitdev_]

AI has turned coders into writers.

The more articulate you are, the better coder you will be.

[Rohit Yadav @rohitdev_]

A lot of things have become possible due to vibe coding.

Features that were earlier not worth investing time on can be shipped fast.

It’s time when software would get more refined.

[Rohit Yadav @rohitdev_]

The bottleneck isn’t coding anymore.

It���s thinking clearly enough to express what you want.

Writing is now the highest leverage skill.

[Rohit Yadav @rohitdev_]

@googleaidevs Some serious prompting would have gone into it for sure otherwise we would have got another generic slop. Great if we can get prompts as well.

[Rohit Yadav @rohitdev_]

Spent an hour debugging a deployment flow.

Built it into a reusable skill.

Now the same problem solves itself in 30 seconds.

At scale: developers publish solutions as skills at a marketplace. AI agents discover and apply them instantly.

Knowledge compounds automatically.

[Rohit Yadav @rohitdev_]

Claude is not AI.

It’s a video game.

[Rohit Yadav @rohitdev_]

It’s an art to get things done by LLMs.

It’s called articulation.