Getting Started for Users

Welcome to Nexus! This guide will help you start using Nexus quickly. Nexus is your organization’s gateway for accessing AI models like GPT-4 and Claude. You don’t need to be a developer to get started.

Getting Started

Sign In to Nexus

Sign in to the Nexus console with your work email.

Learn how to sign in →

Get Your API Key

Create an API key to connect your applications to Nexus.

Get your API key →

Connect a Provider

Use your own provider key (BYOK) or shared billing (PTB).

Set up provider access →

View Usage & Costs

Track your requests, costs, and budget in the dashboard.

Check your dashboard →

---

Accessing the Console

The Nexus console is a web application where you can manage your API keys, view usage, and configure settings. It is your central hub for everything related to your AI usage: creating credentials, monitoring costs, reviewing audit logs, and setting up provider access. The console provides a visual interface that makes it easy to understand what is happening across your projects and organizations without writing any code.

How to Sign In

1. Get your Nexus URL from your administrator (for example: https://nexus.yourcompany.com)
2. Open the URL in your web browser
3. Enter your work email and password
4. If this is your first time, your admin may need to invite you first

Once you sign in, you’ll see your dashboard with an overview of your workspace.

Understanding Your Workspace

Nexus uses a hierarchical structure to organize users, projects, and resources within your organization.

Nexus organizes access in three levels:

• Workspace: Your top-level account (usually one per company)
• Organization: A team or department within your workspace
• Project: A specific application or use case

This hierarchy exists to give you control and visibility at different scales. A workspace represents your entire company’s Nexus deployment. Within that workspace, you create organizations for different teams: maybe one for Engineering, one for Product, and one for Marketing. Each organization can have its own budget, its own classification rules, and its own set of users.

Projects live inside organizations and represent individual applications or use cases. For example, your Engineering organization might have separate projects for your customer support chatbot, your internal documentation assistant, and your code review tool. Each project gets its own API keys, its own usage tracking, and its own cost reporting.

This separation helps you answer questions like “How much is the Marketing team spending on AI?” or “Which application is using the most GPT-4 tokens this month?” It also lets you set different security rules for different use cases. For example, your customer support chatbot can handle confidential data, but your marketing content generator can only handle public information.

Most users work within one organization and one or more projects. Your administrator sets this up for you when they give you access to Nexus.

Example Structure:

Acme Corp (Workspace)
├── Engineering (Organization)
│   ├── Customer Support Bot (Project)
│   │   └── API Key: nxs_support_prod_...
│   ├── Documentation Assistant (Project)
│   │   └── API Key: nxs_docs_prod_...
│   └── Code Review Tool (Project)
│       └── API Key: nxs_code_review_...
│
├── Product (Organization)
│   ├── User Research Assistant (Project)
│   └── Feature Ideation Bot (Project)
│
└── Marketing (Organization)
    ├── Content Generator (Project)
    └── Social Media Assistant (Project)

Each project has its own API keys, budget tracking, and usage reporting. Organizations can set classification ceilings and spending limits that apply to all their projects.

Security Benefits

This workspace structure provides important security isolation. Because API keys are scoped to individual projects, a compromised key can only access that one project’s resources. If someone gets access to your customer support bot’s API key, they cannot use it to access your internal documentation assistant or your code review tool. This limits the blast radius if a key is leaked.

The hierarchy also enables defense in depth through layered security controls. Your workspace might enforce company-wide policies like requiring multi-factor authentication. Organizations can add their own rules, such as classification ceilings that prevent certain teams from handling highly sensitive data. Projects can further restrict access by setting spending limits or rate limits specific to that application.

This means different applications can have different security postures based on their risk profile. Your customer-facing chatbot might handle only public and internal data, while your HR analytics tool can access confidential employee information. If the chatbot is compromised, the attacker still cannot reach the HR data because the projects are isolated. Each layer of the hierarchy adds another security boundary, making it harder for threats to spread across your organization.

Creating an API Key

API keys let your applications talk to Nexus. Each key is tied to a specific project.

When You Need an API Key

An API key is your application’s authentication credential for making requests to Nexus. Think of it like a password, but specifically for software instead of people. Every time your application wants to call an AI model through Nexus, it includes the API key in the request to prove it has permission.

You need an API key if you are building an application that uses AI, running scripts or automation that call AI models, or integrating Nexus into existing software. For example, if you are building a customer support chatbot, a content recommendation engine, or a data analysis tool that uses GPT-4 or Claude, your application needs an API key to send requests to Nexus.

You do not need an API key if you are just viewing dashboards, checking usage reports, or managing settings in the Nexus console. Those activities use your user login credentials. API keys are only for programmatic access: when code, not a person, is making the request.

How to Create an API Key

1. Open the Nexus console and sign in
2. Navigate to your project (select it from the top menu)
3. Click “API Keys” in the sidebar
4. Click “Create New Key”
5. Give it a name (like “My App” or “Production Key”)
6. Copy the key - it starts with nxs_ and you’ll only see it once
7. Save it somewhere safe (like a password manager)

Keeping Your Key Safe

Treat your API key like a password. Never share it in emails, chat messages, or public places. Do not put it in code that you commit to Git or share publicly. If you accidentally commit a key to a public repository, anyone can find it and use it. Instead, store keys in environment variables, configuration files that are not committed to version control, or secure secrets management systems like password managers or cloud secret vaults. If you think your key was exposed, delete it immediately and create a new one. Revoked keys stop working instantly, which prevents unauthorized access.

Why it matters

Your API key identifies your application and controls access to your project. A leaked key can lead to unauthorized usage and unexpected costs.

Setting Up Provider Access

Nexus connects to AI providers like OpenAI (GPT models) and Anthropic (Claude models). Before you can make requests, you need to set up how your organization pays for AI usage. There are two billing models: Bring Your Own Key (BYOK) and Pass-Through Billing (PTB). Both models give you the same AI capabilities, but they differ in who manages the provider relationship and how costs are tracked.

Option 1: BYOK (Bring Your Own Key)

With BYOK, you use your own API key from OpenAI or Anthropic. The provider bills you directly for usage, and Nexus acts as a secure proxy that adds tracking, auditing, and policy enforcement on top of your existing provider relationship.

BYOK is ideal if you already have an OpenAI or Anthropic account, if you want direct billing from the provider to maintain your existing pricing or contract terms, or if your organization requires strict control over which provider credentials are used. Many regulated industries prefer BYOK because it keeps the direct business relationship between your organization and the AI provider, with Nexus adding a governance layer without becoming the financial intermediary.

How to set up BYOK:

1. Get an API key from OpenAI or Anthropic
2. Open the Nexus console and go to your project
3. Click “Provider Keys” in the sidebar
4. Click “Add Provider Key”
5. Select the provider (OpenAI or Anthropic)
6. Paste your key and save it

Nexus stores your provider key securely using encryption at rest. Your provider key never leaves your organization’s Nexus instance: it stays within your deployment boundary and is never sent to external systems. When you make a request, Nexus retrieves the key from secure storage, uses it to authenticate with the provider, and then returns the response to your application. This design ensures that even if someone intercepts the traffic between your application and Nexus, they cannot see your provider credentials.

Option 2: PTB (Pass-Through Billing)

With PTB, your organization manages a shared wallet funded with AI credits. You draw from this wallet when making requests, and your administrator handles the billing relationship with providers. Nexus uses platform-managed credentials to call the AI providers on your behalf and deducts the cost from your organization’s wallet balance.

PTB simplifies billing by consolidating all AI spending into one place. Instead of individual teams managing their own OpenAI or Anthropic accounts, everyone uses the shared wallet. This makes budgeting more predictable and reduces administrative overhead: your finance team sees one line item for AI costs instead of dozens of separate provider bills. PTB also enables features like pre-request cost estimation and budget enforcement. Before forwarding a request to the provider, Nexus reserves the worst-case estimated cost from your wallet. If the wallet does not have enough credits, the request is denied before any charges occur, preventing accidental overspending.

When to use PTB:

PTB works well when your organization provides AI credits as a shared resource, when you want simplified billing with one consolidated statement instead of many individual accounts, or when you do not want to manage your own provider accounts and credentials. It is especially useful for teams that are just getting started with AI and do not yet have established relationships with OpenAI or Anthropic.

How to check your wallet balance:

1. Open the Nexus console
2. Click “Wallet” in the sidebar
3. View your balance and recent transactions

If you run low on credits, ask your administrator to add more to the wallet. Wallet top-ups are managed at the workspace or organization level, so individual users do not need direct access to provider billing systems.

Can You Use Both?

Yes. You can use BYOK for some models and PTB for others, or configure fallback behavior so Nexus tries your BYOK credentials first and falls back to PTB if your personal quota is exhausted. This hybrid approach gives you flexibility: use BYOK when you need guaranteed capacity or specific contract pricing, and rely on PTB as a backup or for non-critical workloads.

Nexus always tries BYOK credentials first if they are available. If the BYOK request fails due to rate limiting, quota exhaustion, or provider downtime, Nexus automatically retries using PTB credentials if your organization has enabled them. This failover happens transparently, so your application does not need to handle the complexity of managing multiple provider keys or retry logic.

The routing logic also enforces security policies during provider selection. For example, if your organization has classified a request as confidential or restricted, Nexus checks whether the selected provider credentials are approved for that classification level. If BYOK credentials do not meet the classification policy but PTB credentials do (or vice versa), Nexus routes the request accordingly. This ensures that sensitive data only reaches providers that meet your organization’s compliance requirements.

Why it matters

Requests cannot reach an AI model until a provider is connected. Choosing the right billing model keeps costs predictable and accountable, while Nexus ensures provider credentials are stored securely and used only within your policy boundaries.

Making Your First Request

Once you have an API key and provider access, you are ready to make requests. Nexus works with the same tools and code you already use for OpenAI or Anthropic. You just need to change your base URL to point to your Nexus instance (instead of api.openai.com) and use your Nexus API key (instead of your OpenAI key). Everything else stays the same. Your existing code, libraries, and tools will work without changes.

Example: Connecting in Plain Terms

Instead of sending requests directly to OpenAI at https://api.openai.com using your OpenAI key sk-..., you will send requests to Nexus at https://nexus.yourcompany.com using your Nexus key nxs_.... Nexus handles the rest: routing to the right provider, tracking costs, and logging for compliance.

Monitoring Your Usage

The Nexus dashboard helps you track AI usage and spending.

Usage Metrics:

• Number of requests you’ve made
• Which AI models you’re using
• How many tokens (words) you’ve used
• Response times and success rates

Cost Tracking:

• Cost per request
• Daily, weekly, and monthly spending
• Breakdown by model and provider
• Budget alerts (if your admin set them up)

Request Details:

• When each request happened
• What you asked and what the AI replied (if tracing is on)
• Any errors or issues
• Classification labels (if you’re using data classification)

How to View Your Dashboard

1. Sign in to the Nexus console
2. Select your project from the top menu
3. Click “Dashboard” to see charts and recent activity
4. Filter by date range to see specific time periods
5. Click on any request to see detailed information

Understanding Costs

Each AI model has different pricing. GPT-4 costs more per request than GPT-3.5, and longer conversations use more tokens and cost more. Image generation and advanced features have separate pricing. Your dashboard shows exactly what you are spending, and you can compare pricing across different models in the model registry to help you choose the most cost-effective option for your use case. If costs seem high, consider using less expensive models when possible, shortening prompts or conversations, or talking to your admin about budget limits.

Why it matters

Regularly reviewing usage helps you catch cost spikes early and stay within your team’s budget.

---

Next Steps

Now that you’re set up, here are some things you might want to explore:

Session Tracking

Group related requests together to track multi-step conversations. This helps you see the full context of AI interactions.

Learn about sessions →

Data Classification

Label your requests with sensitivity levels (public, confidential, etc.). This ensures sensitive data stays protected and meets compliance requirements.

Understand classification →

Audit Logs

View a complete history of who accessed what and when. Audit logs help with compliance, security reviews, and troubleshooting.

Explore audit logs →

Model Selection

Learn how Nexus routes requests to different AI providers and models. Understand fallbacks, cost optimization, and quality tradeoffs.

Learn about routing →