Articles for people who
ship the stack
Original rewrites on React, Node.js, TypeScript and AI — practical notes from the same engineering practice behind our operator software. Article bodies are in English.
Tagged: fine-tuning
Practical notes: What does a model-diffing agent actually discover from a
Operable walkthrough of Practical notes: What does a model-diffing agent actually discover from a: contracts, checks, and drop-in code slots for teams shipping this pattern.
2693 wordsRead articlePractical notes: Small-Model Distillation — Part 2: Off-Policy Soft-Label KD
Operable walkthrough of Practical notes: Small-Model Distillation — Part 2: Off-Policy Soft-Label KD: contracts, checks, and drop-in code slots for teams shipping this pattern.
3831 wordsRead articleSmall Model, Big Leverage: What We Learned Fine-Tuning NVIDIA Nemotron 3.5
Operable walkthrough of Small Model, Big Leverage: What We Learned Fine-Tuning NVIDIA Nemotron 3.5: contracts, checks, and drop-in code slots for teams shipping this pattern.
3861 wordsRead articlePractical notes: Training and Adaptation for Enterprise AI Agents
Operable walkthrough of Practical notes: Training and Adaptation for Enterprise AI Agents: contracts, checks, and drop-in code slots for teams shipping this pattern.
9574 wordsRead articleDiagnosing LLM Output Problems: When to Prompt, Retrieve or Fine-Tune
A symptom-first way to decide whether a weak AI feature needs a better prompt, a retrieval layer or fine-tuning, and why training a model on facts backfires.
1607 wordsRead articleServing a LoRA Fine-Tune Locally: Verify, Fuse and Avoid Silent Failures
Prove a LoRA adapter actually improved a small model, fuse and serve it behind a local OpenAI-compatible API, and catch the failures that return confident wrong output.
2595 wordsRead articleTraining Your Own LLM Judge: From PandaLM and JudgeLM to Prometheus
How finetuned evaluator models are built, from data and rubrics to bias control and meta-evaluation, and a practical recipe for training a domain-specific LLM judge.
8925 wordsRead articleFine-Tune or Call the API? Costing a Document-Extraction Pipeline
A worked cost model for an audit document pipeline shows why model routing beats fine-tuning on price, and when schema accuracy or EU data residency justify owning a model.
3046 wordsRead article
About these articles
Request a 24h estimate
Need the same stack in a production operator layer? Send the brief — estimate within 24 hours.