FAQ

General

What is GridWork HQ? GridWork HQ is a cloud-hosted AI agency operations platform. You connect your tools, build or run pipelines, and get structured outputs — without managing any AI infrastructure yourself.

Is there a free trial? Yes. Every new account starts with a 14-day free trial. No credit card required at signup. After 14 days, your account moves to read-only until you choose a plan.

Is there a self-hosted option? No. GridWork HQ is a cloud-only SaaS product.

What's included in my plan?

Solo: 1,500 credits/month + 1 seat
Team: 6,000 credits/month + up to 15 seats
See Plans & Billing for details.

Billing

How do credits work? Credits are the unit of pipeline execution cost. Each pipeline type has a fixed credit cost (Simple: 1cr, Standard: 3cr, LangGraph agent: 8cr, Full agent+RAG+memory: 15cr). Your plan includes a monthly credit allowance, and you can buy additional packs.

Do credits expire? Monthly included credits reset on your billing anniversary. Purchased credit packs never expire.

What happens when I run out of credits? Pipeline runs are paused until your next billing cycle or until you purchase a credit pack. You receive an email warning when your balance drops below 10% of your monthly allowance.

Can I cancel my subscription? Yes. Go to Settings > Billing > Cancel Plan. Your plan remains active until the end of your current billing period. Data is preserved after cancellation.

How do annual plans work? Pay for 10 months, get 12 months of service. Save 2 months.

Integrations

Why does my integration show "Not configured"? The integration has not been connected yet. Go to Settings > Integrations and follow the setup steps for that integration.

What's the difference between OAuth and API key integrations? OAuth integrations require you to authorize GridWork through the provider's login flow (e.g., "Connect with Google"). API key integrations require you to paste a key from the provider's dashboard into GridWork. Both are stored encrypted and used only for pipeline execution.

Can I use multiple API keys for one integration? No, but you can create separate pipelines that use different integrations.

Pipelines

What's the difference between library and custom pipelines? Library pipelines are pre-built and ready to run — just fill in parameters. Custom pipelines are built by you in the React Flow visual editor, with full control over nodes, connections, and logic.

How do I schedule a pipeline? Open the pipeline, click Schedule, enter a cron expression and timezone, and enable the schedule. See Scheduling for details.

Can I run a pipeline without any integrations? Yes. You can test with local Ollama or BYOK (Bring Your Own Key) OpenAI. Both require zero integrations.

Memory

What's the difference between Facts, Rules, and Learned observations?

Facts — static key-value pairs you write (brand.name, team.lead).
Rules — explicit directives you write ("Always use formal tone").
Learned — AI-proposed observations derived from your patterns (accepted, dismissed, or promoted).

Does GridWork train on my memory or pipeline data? No. GridWork does not use your data to train any model. Memory is used only to enhance your own pipeline prompts.

Security

Where is my data stored? GridWork HQ data is stored in Neon Postgres (US East region). Pipeline outputs and vector embeddings are stored on Fly.io (US East). All data is encrypted at rest and in transit.

Does GridWork train on my data? No. GridWork does not use your pipeline inputs, outputs, or integration data to train any model.

What is BYOK? BYOK (Bring Your Own Key) means you provide your own API key for a service (e.g., OpenAI, Langfuse). GridWork stores it encrypted and uses it only for your pipeline runs. You are billed directly by the provider for usage.

How are API keys stored? All API keys are encrypted at rest using AES-256-GCM. They are decrypted only when needed for pipeline execution.

Technical

What LLMs can I use?

OpenAI (GPT-4, GPT-4 Turbo, GPT-3.5 Turbo) via BYOK
Ollama (local LLM, self-hosted)
Managed LLM add-on (GridWork's API key)

Can I use local models? Yes. Configure Ollama in Settings > Integrations > Ollama and point it to your local instance.

What's the output format? Pipeline outputs are stored as markdown. You can view, edit, and export them from Reports.

Can I export my data? Yes. You can export individual outputs as PDF or markdown from the Reports page.