soasme

---
name: small-model-finetuning
description: Build, evaluate, fine-tune, quantize, and deploy small language models for narrow production tasks using an eval-first dataset factory workflow.
version: 1.0.0
language: en
tags:
  - small-language-models
  - fine-tuning
  - qlora
  - lora

soasme

The New Engineering Teams

How AI coding agents change the way teams should work

---

Opening

Most engineering orgs were designed for one constraint:

soasme

AI Engineering: LLMs, RAG, MCP, Agents, Fine-Tuning, and Quantization

This post covers the core concepts of AI engineering — not machine learning research, but the practical discipline of taking pre-built models and shipping them into real products. By the end, you'll have a working mental model of LLMs, RAG, MCP, agents, fine-tuning, and quantization, and understand how they fit together.

---

AI Engineering vs. Machine Learning

These two disciplines are often conflated, but they're distinct.

soasme

AI Engineering: A Complete Overview of LLMs, RAG, MCP, Agents, Fine-Tuning, and Quantization

Artificial Intelligence has rapidly evolved from a niche research field into one of the most transformative technologies in software engineering. Today, developers across Android, backend, web, DevOps, and cloud engineering are increasingly expected to understand AI systems and how to integrate them into real-world products.

This article walks through the core concepts behind modern AI engineering — including Large Language Models (LLMs), Retrieval-Augmented Generation (RAG), MCP (Model Context Protocol), tools, embeddings, and the architecture of AI-powered systems.

---

Machine Learning vs AI Engineering

soasme

import logic from 'logicjs';
var or = logic.or,
	and = logic.and,
	not = logic.not,
	eq = logic.eq,
	run = logic.run,
	lvar = logic.lvar,
	between = logic.between

var g1_rule = `

soasme

Core Problem: Users need a frictionless way to log daily actions and track their impact without overwhelming complexity.

Invariants for Success:

1. Users can log actions in under 10 seconds.
2. The feed clearly shows the relationship between actions and outcomes.
3. The default view prioritizes today's activities.
4. Reflection is optional and unobtrusive.

Must-Have Outcome of the MVP: A single screen that allows users to log actions quickly, view a feed of their logs, and see metrics associated with those actions.

soasme

soasme

{ pkgs ? import <nixpkgs> {}}:

with pkgs; mkShell {
  # Include C++ headers for regular clang calls:
  NIX_CFLAGS_COMPILE = lib.optionals stdenv.isDarwin [
    "-I${lib.getDev libcxx}/include/c++/v1"
  ];

  nativeBuildInputs = [
    ...

soasme

soasme

---- MODULE VirtualMachine ----

VM_Version == "1.0.0"

LOCAL INSTANCE Sequences
LOCAL INSTANCE Integers
LOCAL INSTANCE TLC

CONSTANT PC
CONSTANT SUBJ