Train AI Models · Fine-tuning & Post-training

Post-train a frontier model on your domain.

When the problem is language and judgment, we take a strong base model and post-train it on your data, voice, and format — supervised fine-tuning and preference optimization — and prove the gain on held-out evals before it ships.

The pipeline

How a base model becomes yours.

A disciplined sequence, with an eval gate before anything ships.

1

Base model

Start from the right frontier or open model for the task, licensing, and deployment target — chosen, not defaulted.

2

Supervised fine-tuning

Teach the model your task, format, and tone from curated input-output pairs — the fastest path to a specialized model.

3

Preference optimization

Align behavior to human judgment with RLHF or DPO — rewarding the answers your experts prefer, penalizing the ones they don't.

4

Evaluation

Prove the tuned model beats the base on a held-out benchmark for quality, safety, and regression before it goes near production.

5

Distillation & serving

Compress into a smaller, cheaper model where you can, then serve behind an API with routing and monitoring.

Techniques

The methods we reach for.

Supervised fine-tuning (SFT)

Curated demonstrations teach task, format, and domain language — the highest-leverage first step for most specialization.

Instruction dataLoRA / fullFormat control

RLHF

Reinforcement learning from human feedback trains a reward model on preferences, then optimizes the policy against it.

Reward modelPPOPreference pairs

DPO

Direct preference optimization skips the separate reward model — simpler, stable, and often enough.

Preference pairsNo reward modelStable

LoRA / PEFT

Parameter-efficient tuning adapts a large model by training small adapters — cheaper, faster, and easy to swap.

AdaptersLow VRAMSwappable

Distillation

Transfer a big model's behavior into a small one to cut latency and cost while keeping most of the quality.

Teacher → studentLatencyCost

Continued pretraining

When your domain language is genuinely unusual, keep pretraining on your corpus before task tuning.

Domain corpusVocabularyFoundational
Choose the right tool

Prompt, retrieve, or fine-tune?

Fine-tuning is powerful and often unnecessary. Here is how we decide.

ApproachBest forCost to changeWatch out for
Prompt engineeringBehavior a strong model can already do with the right instructions and examples.MinutesGets fragile as prompts grow; token cost on every call.
Retrieval (RAG)Knowledge that changes or must be cited — facts live outside the model.HoursRetrieval quality is the ceiling; won't change style or skill.
Fine-tuningA consistent skill, format, or tone the base model doesn't have — proven at scale.Days to weeksNeeds quality data and re-training as the world moves.
Honest advice

For most knowledge problems, retrieval is cheaper, more current, and easier to audit than training. We fine-tune when a task needs a skill or consistency the base model lacks — and we tell you when it doesn't, instead of selling a training project you don't need.

Get started

Bring us a hypothesis. Leave with a system.

Tell us what's eating your team's time. We'll give you an honest read on whether AI is the right tool — and if it is, a scoped v1 with a timeline and cost.