Build a Perplexity-style answer engine

What you’re building: an answer engine that searches the web, reads the sources, and answers with citations, backed by a vector index.

Primitives you’ll use: Organization, Sandboxes, Computers, Agents, Qdrant vectors, Deployments, Custom domains

What you’re building

An answer engine has three jobs: find sources, read them, and answer with citations. Each job fits a different MIOSA primitive.

Use a Computer for the browsing step, because a real desktop browser handles the pages a plain HTTP fetch cannot. Use a Sandbox to extract and clean text, chunk it, and write the pipeline code. Use managed Qdrant to store embeddings so the same sources power follow-up questions, and Postgres for documents and users.

The result is a pipeline you can run live for a query and reuse for the corpus, with the browsing, the compute, and the index all on MIOSA.

What you need on MIOSA

Architecture

Step 1: Create an account and an API key

Install the CLI, sign in, and mint a key for your product. The secret is printed once; store it as an environment variable, never in code.

npm i -g @miosa/cli
miosa login
miosa whoami
miosa api-key create answer-engine-key --preset agent     # prints the secret once
export MIOSA_API_KEY="msk_u_..."

Both SDKs read MIOSA_API_KEY from the environment, so a server or a worker needs nothing else. See API keys for scopes and how to create keys in the console.

Step 2: Create the resources

The sandbox gets QDRANT_URL and QDRANT_API_KEY for the managed instance.

Step 3: Wire the agent loop

See Qdrant vectors for the client and collection setup.

Step 4: Preview

A Preview is a throwaway public URL for a port inside the machine. Start the app as a background process so it survives the CLI disconnecting, then expose its port.

Open the exact Preview URL MIOSA returns.

Step 5: Deploy

Publish the machine to an immutable Deployment. Give it the source machine, the path the app lives at inside it, and the command that serves it.

Publishing the same deployment name again creates a new immutable version and keeps the same URL. Roll back with miosa deploy rollback answer-engine --to <version-id>, and inspect a live app with miosa deploy logs answer-engine -n 100. See Publishing and Rollback.

Step 6: Custom domain

Attach a domain your customer owns to the deployment. The platform URL keeps working while DNS propagates.

miosa deploy domain-add answer-engine app.example.com
miosa deploy domains answer-engine          # shows the DNS record and verification target
miosa deploy domain-verify answer-engine <domain-id>

Copy the exact DNS record MIOSA shows into the customer’s DNS provider; for a subdomain it is normally a CNAME. Once verified and TLS is active, MIOSA routes the domain to the deployment’s active version. In code, read the URL from the publish response instead of building it: the SDK method is miosa.deployments.domains.add(deploymentId, { domain }) (TypeScript) or miosa.deployments.domains(deployment_id).add(domain=...) (Python). See Domains.

Costs and limits

Everything bills while it runs: machines (sandboxes and computers) by the second, managed databases while running, and deployments when they serve. Pausing a machine stops compute billing but keeps disk. Set an --idle-timeout or a --ttl so a forgotten machine cannot run forever, and cap spend with rate limits and per-workspace quotas.

• Pricing - plans, the published rate card, and the free Developer grant.
• Sizing and limits - machine sizes and what each costs to run.
• Timeouts and auto-stop - --ttl, --idle-timeout, and pause/resume.

Next steps