AI Help Resources
Everything you need to understand AI models — from zero to building your own.
Quick Concepts
Before diving in, here are the core terms you'll encounter. Each explained in one sentence.
AI Model
A program trained on data to recognize patterns and make predictions. Think of it as a brain that learned from millions of examples.
Training Data
The examples a model learns from. More quality data = smarter model. Like a student studying textbooks.
Training
The process of feeding data into a model so it learns patterns. Requires significant compute power (GPUs).
Fine-Tuning
Taking a pre-trained model and teaching it something specific. Like a doctor who first went to med school (pre-training) then specialized (fine-tuning).
Prompt Engineering
Writing instructions that get the best output from an existing model. No training needed — just better communication.
The AI Stack
Infrastructure → Base Model → Fine-Tune → Prompts → Application. Most teams work on the top layers, not the bottom.
Learning Tracks
Structured paths from beginner to advanced. Follow them in order, or jump to what interests you.
AI Fundamentals
Understand what AI models are, how they work, and the key concepts behind them.
What is an AI Model?
The brain analogy — what a model actually is, how it stores knowledge, and why it matters.
How AI Models Learn
Training, datasets, weights, and the process of teaching a model to understand patterns.
Types of AI Models
LLMs, image models, code models, embeddings — and when to use each one.
Building vs Using
The critical difference between building an app on top of AI and building the AI model itself.
App on Top vs Model Itself
Why most companies build applications that use existing models — and when you actually need your own.
Fine-Tuning Explained
Taking an existing model and specializing it for your use case — the middle ground.
Prompt Engineering vs Training
When to write better prompts, when to fine-tune, and when to train from scratch.
The AI Stack
Infrastructure → Model → Fine-tune → Prompt → Application — understanding each layer.
Build Your Own Model
From data collection to deployment — how to train, fine-tune, and ship a custom model.
Collecting Training Data
Where data comes from, how to clean it, and how much you need.
Training Your First Model
Step-by-step: from dataset to a working model you can talk to.
Fine-Tuning with Codmir
Use Codmir's platform to fine-tune open-source models on your own data.
Deploying Your Model
Serving your model via API — on your cluster, on cloud, or hybrid.
AI on Codmir
How Codmir uses AI under the hood and how you can leverage it in your projects.
How Codmir Uses AI
Chat, code generation, task execution, review chains — the AI systems inside Codmir.
AI Modes & Presets
Supervised, Collaborative, Autonomous — choosing the right mode for your workflow.
Connecting Your Own Models
Use Codmir's control plane to route tasks to your own locally-hosted models.
Ready to build?
Start with the fundamentals or jump straight into building your own model on Codmir's cluster-first platform.