Field notes on post-training.
Practical, vendor-neutral guides from engineers who fine-tune models for a living. No fluff, no gated PDFs.
Fine-tuning vs RAG: how to choose (and when to use both)
A practical framework for choosing between retrieval-augmented generation and fine-tuning, with cost, accuracy and maintenance trade-offs.
SFT vs DPO vs RLHF: choosing a post-training method
What each post-training method teaches a model, how much data it needs, and how to sequence them for an enterprise task.
How much does it cost to fine-tune an LLM?
A line-by-line breakdown of data, compute, evaluation and serving costs, with three worked examples from 8B to 70B.
LLM distillation: a practical guide for enterprise teams
How to move frontier-model behavior into a small, private model — data generation, filtering, training and pitfalls.
How to evaluate a fine-tuned LLM before it reaches production
Golden sets, LLM-as-judge calibration, regression gates and the statistics you need to trust a result.
Fine-tuning glossary
Plain-English definitions of the post-training terms that come up in every project.
Bring us the task. We'll bring the model.
A 45-minute technical briefing with a senior research engineer. You leave with a feasibility read, a data plan and a fixed quote — whether or not you hire us.
Book a technical briefing →