Another Dev's Two Cents

Nish Tahir

https://nishtahir.com

Notes on - Why do LLMs freak out over the seahorse emoji?

A fantastic deep dive into the seahorse emoji phenomena[1] was recently published by Theia[2]. It's engaging, well presented and worth reading. The post presents a case using meta-llama/Llama-3.3-70B-Instruct. However I wanted to verify this behavior with smaller models which unsurprisingly

How LLM Structured Decoding works

Last week I happened to be in a discussion that involved getting an LLM to generate JSON reliably. A major frustration expressed was that no matter how much they tried the model would often fail to follow instructions during generation. I pointed out that most major vendors support some variant

Notes on the phi-4-reasoning Technical Paper

Phi-4 Reasoning Microsoft recently released the phi-4 reasoning model as well as its technical report. They explore using supervised finetuning as well as synthetic dataset curation to train phi-4 a 14B parameter model to compete with and often outperform significanly larger models such as DeepSeek-R1-Distill-

Writing a backup service for Ubuntu in Rust

While performing some routine maintenance of my Ubuntu server, $ docker compose pull $ docker compose up -d I ran into an interesting situation where nginx-proxy-manager failed to serve my blog properly, I wasn't sure what the issue was but these types of issues are usually solved with

Adding Gaussian Noise to a signal in Python

Gaussian noise is data that is added to a signal in order to introduce a distortion. The data follows a Gaussian/Normal distribution. It's a well understood distribution often used to introduce noise to training data as an augmentation technique. Generating noise to add to a signal is

Building an Ubuntu service for my Antec Flux Pro

I recently put together a new PC build. It's primarily aimed at machine learning so I wanted to get a case that had really great airflow. I asked around and was directed to a list put together by GamersNexus[1] highlighting their top cases of the year. The

Notes on the DuckDB UI

DuckDB recently shipped with a new local UI package[1]. It allows users to visualize DuckDB databases as well as run SQL queries through embedded notebooks. You can start the UI by running duckdb -ui in the terminal which starts an http server on port 4213 by default. $ duckdb -ui

Vector Projection

Vector projection is the process of projecting one vector onto another. In simple terms, it is the adjacent side of a right triangle with the original vector as the hypotenuse.

Vector Normalization

Vector normalization is the process of scaling a vector to have a magnitude of 1 or unit length. In other words, we want to change the length of the vector to 1 while keeping the direction the same.

The Value of Expertise

I've been reading a lot about the impact that AI[1] has had and will have on the software industry. Some herald the death of Software Engineering as a profession[2], others claim it's a full-on paradigm shift in our work[3], while some hold