Salman Quazi

When you type a sentence into an LLM, the model never sees your letters.

In 1964, a Canadian scientific expedition landed on Easter Island — Rapa Nui — a volcanic speck in the South Pacific, 2,300 miles from the nearest continent.

In October 2016, the Nobel Committee awarded the Prize in Physiology or Medicine to a 71-year-old Japanese cell biologist named Yoshinori Ohsumi.

Every token you’ve ever received from an LLM was generated one at a time.

Every time you run a serverless function on AWS Lambda, execute a container on Google Cloud Run, or spin up a GitHub Actions workflow, your code runs on a physical machine shared with hundreds of other tenants.

Tool call hallucination comes in three flavors.

When AI agents first started multiplying, every vendor shipped their own integration approach.

There’s a famous chart that shows a near-perfect correlation between the per capita consumption of mozzarella cheese in the United States and the number of civil engineering doctorates awarded each year.

There’s a tension at the heart of software that functional programming makes explicit: side-effects are what make programs useful, yet they’re also what make programs hard to reason about.

Software supply chain attacks don’t exploit your code—they exploit your trust.

Give a language model a few examples of a pattern — say, foo → FOO, bar → BAR, baz → — and it completes the sequence correctly without retraining.

The container runtime is what you interact with.

In a previous post, we looked at post-training as a category — SFT, RLHF, DPO — and contrasted it with in-context learning.

If you’ve spent any time working with large language models, you’ve probably encountered two very different ways of getting them to do what you want: post-training and in-context learning.

In the first post of this series, we looked at how special tokens like <|im_start|> and <|im_end|> form the structural grammar of LLM conversations, and how the newer Harmony format extends this with tokens like <|call|> and <|return|> for tool invocation…

The AI tooling landscape has fractured into a bewildering number of SDKs, frameworks, and agents – each claiming to be the right way to build with large language models.

Andrej Karpathy recently published microGPT – a complete GPT implementation in 200 lines of pure Python with zero dependencies.

In the previous posts in this series, we established that when you give an LLM a list of function tools, the model must interpret JSON schemas at inference time and produce structured output that conforms to them.

When you connect multiple MCP servers to a coding agent like Claude Code, something specific happens to the model’s context window at every step of the reasoning loop.

Every AI agent framework — LangChain, LlamaIndex, Microsoft Agent Framework, CrewAI — wraps the same idea.

I was in my late thirties when I started paying attention to my health – not in a vague, someday kind of way, but urgently.

In the previous post, we established that built-in tools like code_interpreter and web_search outperform custom function tools because they’re in-distribution – the model was trained on their exact invocation patterns during post-training.

When you give an LLM a list of tools, two things can happen.

If you’ve ever looked at the raw token stream behind a ChatGPT conversation, you’ve seen things like <|im_start|>, <|im_end|>, and <|im_sep|>.

Large language models are great at calling tools — but when a task requires chaining several tools together, the model ends up shuttling data back and forth, burning tokens on intermediate results it doesn’t need to see. mcp-compose fixes this by letting…

It’s been a relief to move away from the bloat of WordPress.

As most of you probably already know WebSocket provides full-duplex communication over a single TCP connection. .NET 4.5 added support for WebSockets as part of the BCL.

Most software built today have a notion of a client and a service.

Sebastian Junger (pronounced Younger) was on Bill Maher last night.

In most computer science programs, machine learning is usually a graduate level course.

Update (3/31/2013 - 11:41 PM PST): This article has been updated!

One of the first things I did when I started this blog was to change the way static files are cached by IIS to improve site speed.

Came across this wonderful quote from Jeff Hammerbacher: “The best minds of my generation are thinking about how to make people click ads”.

There are several ways to extend the behavior of a class.

Monitoring your Azure services couldn’t get easier.

Here’s a cool technology you might not have heard of: scriptcs.

We are at the finale of this three part series.

I’ve always been a junkie when it comes to markup validation.

Last time we looked at what the MVVM pattern was, and how it is used today in XAML-based applications.

There seems to be two types of developers out there.

In this series I will be discussing how the MVVM pattern can be used for almost any type of web application.

If you or your kids want to learn how to program - you can start here: https://code.org/.

So a cool new feature in .NET 4.5 is the ability to create anonymous methods that are asynchronous (async).

When dealing with .NET tasks – we often want to timeout the operation if it does not complete within a certain period of time.

Before I go into the technical details of this article.

If you’ve ever worked on an intranet site, you often want to know, programmatically, if a client that’s accessing your website has the ability to automatically login to your site (Integrated Windows Authentication).

You can drastically improve your PowerPoint presentations by simply leveraging pptPlex!

The MultiView control, Wizard control or even a custom panel-based control is very useful to collect large sets of data in wizard form; and it’s often useful to give visual ques to the user as to which step he/she is in.

I presented a topic on jQuery at our campus web publishers meeting.

Generally all programmers are lazy–and specially when it comes to documentation because it takes so much time and doesn’t necessarily produce any material benefit.