Home

Hi! I’m Alex. Here I write about topics that interest me, including MLOps/ML/AI, software engineering, and working in tech in general.

alex000kimREMOVETHIS@gmail.com

Posts

Jul 25, 2025How to Set Up Work and Personal Git Profiles
Step-by-step guide to configuring separate Git profiles for work and personal projects using SSH keys and conditional includes.
Feb 2, 2025US Tariffs, DeepSeek and OpenAI
Comparing how DeepSeek R1 and OpenAI O3 reason about the economic impact of US-Canada tariffs.
Jan 11, 2025Orchestrating LLM Fine-tuning on Kubernetes with SkyPilot and MLflow: A Complete Guide
A complete guide to orchestrating LLM fine-tuning on Kubernetes using SkyPilot for resource management and MLflow for experiment tracking.
Jan 4, 2025Kubernetes Mental Model
A comprehensive mental model for understanding Kubernetes concepts, from clusters and pods to services and storage.
Nov 24, 2024Intro to SLURM for ML Practitioners
A practical guide to SLURM for ML practitioners covering fundamental concepts, key commands, and distributed multi-node training.
May 5, 2024Experiments with OpenAI's Function Calling
Demonstrating OpenAI's function calling API to convert natural language queries into SQL against the Northwind database.
Sep 8, 2023Fine-Tuning Large Language Models with a Production-Grade Pipeline
This post describes a production ML pipeline for fine-tuning large language models using DVC, SkyPilot, HuggingFace Transformers, and quantization techniques.
Aug 16, 2023Trying to Understand Something Difficult? Minimize the Number of Attempts!
Why minimizing context switches and staying in deep focus leads to faster understanding of difficult concepts.
Aug 10, 2023ML experiments in the cloud with SkyPilot and DVC
How to use SkyPilot and DVC to run reproducible ML experiments in the cloud without infrastructure headaches.
Aug 2, 2023Why Sales Engineers Exist
The role of sales engineers in B2B startups and why they're essential for selling enterprise software.
Jul 31, 2023Don’t know what to do next? Teach!
Why teaching is the most powerful way to learn, backed by the Feynman Technique and the Protege Effect.