# Inquir Compute

> Inquir Compute is a cloud platform for your applications. Deploy from Dockerfiles or container images and run web apps, APIs, workers and databases in one project. Connect services over a private network, inspect logs and CPU/RAM metrics, configure alerts, add replicas and roll back releases. Functions in Node.js, Python and Go, API routes and pipelines are part of the same platform. Cron schedules are triggers within pipelines, not a separate service.

Public marketing and documentation live on **https://inquir.org** (English at unprefixed paths, Russian at `/ru/…`). Authenticated product UI under `/app`, `/functions`, and similar paths is not intended for indexing; see `/robots.txt`.

## Applications

- [Home](https://inquir.org/): Container applications — any Dockerfile or image as a long-lived service with health-checked releases, preview URLs, one-click rollback, a private network between services (`<slug>.apps.internal`), Postgres and Redis templates on persistent volumes, raw TCP ports, replicas and custom domains with TLS. Deploys from GitHub repositories are in early access, on request.
- [Tutorial: deploy an application](https://inquir.org/docs/deploy-an-application): From a Dockerfile to production — create, deploy, health check, promote, add a Postgres and a domain, roll back.
- [Deploy recipes](https://inquir.org/docs/recipes): Copy-paste Dockerfile + inquir.yaml for a React (Vite) site, a static HTML site, a Node.js (Express) API and a Python (FastAPI) API, the deploy commands, and adding Postgres.
- [Tutorial: Node.js service with Postgres](https://inquir.org/docs/node-postgres): A Node.js API with a pg pool, SQL migrations at startup under an advisory lock, a second service on the private network, expand/contract schema changes, rollback.
- [Container applications](https://inquir.org/docs/apps): Long-lived container services — Dockerfile or image releases, health checks, preview vs production URLs, TCP ports, volumes and backups, custom domains, promote and rollback.
- [Stacks: inquir-compose.yaml](https://inquir.org/docs/compose): Several services from one file, deployed with one command (preview, enabled per workspace) — services as folders with their own inquir.yaml or inline blocks, inputs, server-generated secrets, references such as `{{services.db.privateHost}}`, dependsOn start order, deploy from a folder, GitHub or the catalog, update, down/delete, rotate and pull.
- [Custom domains](https://inquir.org/features/custom-domains): Bring your own domain — TXT verification and an automatic TLS certificate.
- [CLI](https://inquir.org/features/cli): The npm-installed CLI (`@inquir/compute-cli`) for container applications and functions — create, deploy, promote, connect, roll back.
- [Telegram and MAX bot hosting](https://inquir.org/bots): Bots as always-on container applications with Postgres and Redis templates, an HTTPS address for webhooks, schedules, logs and volume backups. Ready bot examples (Telegram in Python and Node.js, MAX, an AI assistant on any OpenAI-compatible API) with step-by-step deploy guides: https://github.com/Inquir-search/inquir-examples; the guide: https://inquir.org/docs/apps#telegram-bot.
- [Deploy from Claude Code, Cursor and Codex](https://inquir.org/agents): `inquir mcp` is a stdio MCP server — deploy, invoke, deploy status, invocation history and rollback as agent tools.
- [On-premise](https://inquir.org/on-premise): The platform on your own servers for integrators and hosting providers — multi-node control plane, runners and data inside your perimeter; starts with a pilot.
- [Plans](https://inquir.org/pricing): A plan is required to run workloads; a new account gets Free for 14 days automatically after confirming its email (no card, once per account), and a plan covers all the workspaces its owner owns. Simple and Pro are paid plans, Enterprise terms are individual; plans differ only in resources.

## Docs and overview

- [Documentation](https://inquir.org/docs): Product help and reference.
- [Functions](https://inquir.org/features/lambdas): Serverless functions for HTTP requests, webhooks and pipeline steps, including scheduled pipelines, alongside container applications in the same workspace.
- [Platform spec](https://inquir.org/platform-spec): Canonical product facts and limits.
- [Features](https://inquir.org/features): Feature index.
- [Changelog](https://inquir.org/changelog): Release notes.
- [Blog](https://inquir.org/blog): Product updates and guides; individual post URLs are listed in `https://inquir.org/sitemap-blog.xml` (separate from `sitemap.xml`).
- [Templates](https://inquir.org/compute-marketplace): App templates (PostgreSQL, Redis, Nginx) on the project canvas and function starters.

## Documentation reference

- [Quick start](https://inquir.org/docs): Hub and getting-started guide.
- [Use cases (docs)](https://inquir.org/docs/use-cases): In-product playbooks for AI agents, webhooks, scheduled pipelines, and REST APIs.
- [Tutorial: deploy an application](https://inquir.org/docs/deploy-an-application): From a Dockerfile to production — create, deploy, health check, promote, add a Postgres and a domain, roll back.
- [Deploy recipes](https://inquir.org/docs/recipes): Copy-paste Dockerfile + inquir.yaml for a React (Vite) site, a static HTML site, a Node.js (Express) API and a Python (FastAPI) API, the deploy commands, and adding Postgres.
- [Tutorial: Node.js service with Postgres](https://inquir.org/docs/node-postgres): A Node.js API with a pg pool, SQL migrations at startup under an advisory lock, a second service on the private network, expand/contract schema changes, rollback.
- [Stacks: inquir-compose.yaml](https://inquir.org/docs/compose): Several services from one file, deployed with one command (preview, enabled per workspace) — services as folders with their own inquir.yaml or inline blocks, inputs, server-generated secrets, references such as `{{services.db.privateHost}}`, dependsOn start order, deploy from a folder, GitHub or the catalog, update, down/delete, rotate and pull.
- [Tutorial: write and deploy a function](https://inquir.org/docs/deploy-a-function): A handler in the browser or with the CLI, run, deploy, a gateway URL, traces.
- [Tutorial: alerts for pipelines](https://inquir.org/docs/pipeline-alerts): Alert rules for pipeline step runs and human-gate pauses — conditions, Slack and webhook channels, cooldowns, automation.
- [Tutorial: human in the loop](https://inquir.org/docs/human-in-the-loop): The Human gate node, approve and question modes, the inbox, resuming runs from the API, alerting approvers.
- [Functions (docs)](https://inquir.org/docs/functions): Handler model, configuration, warm containers, and test panel.
- [Browser Web IDE](https://inquir.org/docs/editor): Monaco editor, file tree, run, deploy, and logs.
- [Runtimes](https://inquir.org/docs/runtimes): Node.js 22, Python 3.12, and Go 1.22 handler conventions; one slim Debian bookworm image per runtime.
- [inquir.yaml build configuration](https://inquir.org/docs/inquir-yaml): One file for Node.js, Python and Go functions on every deploy surface — OS packages (`build.system`: presets such as ffmpeg, or Debian packages such as Chromium) and language packages (`build.dependencies`: `name@version` for Node.js, a pip requirement for Python, `module@version` for Go; rolling out per workspace). Function layers are removed: declare a function's packages here.
- [Observability](https://inquir.org/docs/observability): Traces, logs, execution history, and SDK spans.
- [API Gateway](https://inquir.org/docs/gateway): HTTP routing, auth modes, rate limits, and event formats.
- [Pipelines](https://inquir.org/docs/pipelines): Step and graph workflows with cron schedule or event triggers, and async jobs.
- [Environment variables and secrets](https://inquir.org/docs/env-secrets): Per-function env config and sensitive values.
- [API Keys](https://inquir.org/docs/api-keys): Gateway API key auth, rotation, and CLI key import.
- [VS Code extension](https://inquir.org/docs/vscode): Deploy, invoke, and log tailing from the editor.
- [Keyboard shortcuts](https://inquir.org/docs/keyboard): Browser IDE hotkey reference.

## Function and platform features

- [API Gateway](https://inquir.org/features/api-gateway): Public HTTP routing and invocation.
- [Pipelines](https://inquir.org/features/pipelines): Workflow composition, including scheduled runs with cron triggers.
- [Runtimes](https://inquir.org/features/runtimes): Language and execution environment overview.
- [Observability](https://inquir.org/features/observability): Logs, metrics, and debugging.
- [Secrets](https://inquir.org/features/secrets): Secrets and environment variables.
- [Warm containers](https://inquir.org/features/hot-containers): Warm execution model for functions.
- [Container runtime](https://inquir.org/features/container-runtime): Container-based functions.
- [Multi-tenant routing](https://inquir.org/features/multi-tenant-routing): Tenant-aware routing.
- [Browser deploy](https://inquir.org/features/browser-deploy): Edit and deploy from the browser.

## Compare and guides

- [Platform comparison](https://inquir.org/compare): Side-by-side positioning vs Lambda, Vercel, Workers, Trigger.dev, Inngest, and Modal.
- [Use cases](https://inquir.org/use-cases): Scenario-based overview for AI agents, scheduled pipelines, webhooks, background jobs, and APIs.

## Pillar 1: AI Agent Backend

- [Serverless for AI agents](https://inquir.org/serverless-for-ai-agents): Topic hub — tool backends, pipelines, scheduled agents, human-in-the-loop.
- [AI agent backend use case](https://inquir.org/use-cases/ai-agent-backend): Tool HTTP routes, secrets off the model path, async jobs, warm pools for tight tool loops, and multi-step agent pipelines with per-step retries and approval gates.
- [AI agent tools API](https://inquir.org/ai-agent-tools-api): Authenticated serverless endpoints for LLM tools and function calling — gateway API key auth, per-tool secrets, isolated containers, orchestrator dispatch.
- [LLM pipelines use case](https://inquir.org/use-cases/llm-pipelines): Staged retrieval, moderation, tool calls, and summarization as observable pipeline steps.
- [Python serverless functions](https://inquir.org/python-serverless-functions): Python 3.12 with numpy, pandas, scikit-learn for AI agent tools and ML inference.
- [Serverless ML inference](https://inquir.org/serverless-ml-inference): Run ONNX models on Python 3.12 (glibc) — declare onnxruntime/scikit-learn in inquir.yaml build.dependencies, load the model once on cold start, CPU only.
- [Node.js serverless functions](https://inquir.org/nodejs-serverless-functions): Node.js 22 with native modules for agent tool backends.
- [Go serverless functions](https://inquir.org/go-serverless-functions): Go 1.22 with standard net/http handlers and CGO support.
- [Docker serverless functions](https://inquir.org/docker-serverless-functions): Docker-isolated serverless without custom Dockerfiles — managed Node.js, Python, Go runtimes behind one gateway.

## Pillar 2: Serverless Background Jobs

- [Long-running serverless jobs](https://inquir.org/long-running-serverless-jobs): Work that outlasts HTTP gateway timeouts — chained pipeline steps (up to 24 h each), fan-out, and where provider time limits live.
- [Serverless async jobs](https://inquir.org/serverless-async-jobs): Queue work from HTTP handlers, webhooks, or cron without a separate worker server; BullMQ and SQS operational comparison.
- [Background jobs with retries](https://inquir.org/background-jobs-with-retries): Fast HTTP 202 handoff, per-step pipeline retries with backoff, idempotent handlers, and searchable failure history—no Redis or worker retry loops.
- [Background jobs for Vercel](https://inquir.org/background-jobs-for-vercel): Keep Next.js on Vercel; move async work to Inquir serverless jobs.
- [Background jobs for Next.js](https://inquir.org/background-jobs-for-nextjs): Compare after(), Server Actions, and Route Handlers vs Vercel limits—enqueue async work on Inquir pipelines without timeout hacks.
- [Inngest alternative](https://inquir.org/inngest-alternative): One function catalog for HTTP, webhooks, and scheduled or background pipelines.
- [Trigger.dev alternative](https://inquir.org/trigger-dev-alternative): Hosted durable workflows versus Inquir pipelines and jobs.
- [BullMQ alternative](https://inquir.org/bullmq-alternative-serverless): Redis worker queues vs serverless pipeline steps without queue infrastructure.
- [CSV processing use case](https://inquir.org/use-cases/csv-processing): Chunked async import with idempotent upsert and progress tracking.
- [PDF processing use case](https://inquir.org/use-cases/pdf-processing): Async document generation and extraction with object storage.

## Pillar 3: Scheduled Pipelines & Webhooks

- [Serverless webhook processor](https://inquir.org/serverless-webhook-processor): HMAC verification, fast ACK, idempotency for provider redeliveries, and async handoff with step retries for Stripe, GitHub, Slack, Shopify.
- [Stripe webhooks use case](https://inquir.org/use-cases/stripe-webhooks): Raw body HMAC, idempotency key, fast ACK, async fulfillment.
- [GitHub webhooks use case](https://inquir.org/use-cases/github-webhooks): timing-safe HMAC, event routing, delivery ID dedup, CI/CD triggers.
- [Slack webhooks use case](https://inquir.org/use-cases/slack-webhooks): 3-second SLA, response_url async pattern, replay-attack protection.
- [Webhooks on your own domain](https://inquir.org/webhooks-on-your-own-server): Stable custom-domain ingress for webhook providers on managed Inquir.
- [Scheduled pipelines guide](https://inquir.org/serverless-scheduled-jobs): Validated cron expressions on pipeline triggers, execution history, retries, and shared secrets.
- [Cron jobs on a VPS](https://inquir.org/cron-jobs-on-vps): Migrate raw crontab to managed scheduled pipelines.
- [Scheduled data sync](https://inquir.org/use-cases/scheduled-data-sync): Incremental watermark cursor sync with idempotent upsert.
- [Nightly ETL](https://inquir.org/use-cases/nightly-etl): Extract, transform, load as pipeline steps — without Airflow.

## Platform comparisons

- [AWS Lambda alternative](https://inquir.org/aws-lambda-alternative): Positioning vs AWS Lambda.
- [Vercel Functions alternative](https://inquir.org/vercel-functions-alternative): Positioning vs Vercel.
- [Cloudflare Workers alternative](https://inquir.org/cloudflare-workers-alternative): Positioning vs Workers.
- [Modal alternative](https://inquir.org/modal-alternative): Positioning vs Modal.
- [Supabase Edge Functions alternative](https://inquir.org/supabase-edge-functions-alternative): Deno V8 isolates vs container-backed Node.js, Python, Go.
- [Railway alternative](https://inquir.org/railway-alternative-serverless-functions): Container apps with databases and private networking, plus functions and pipelines in the same workspace.
- [Render alternative](https://inquir.org/render-alternative-background-jobs): Container apps and background workers, plus pipeline steps with retries in the same workspace.
- [Fly.io alternative](https://inquir.org/fly-io-alternative-serverless-functions): Container apps with volumes, TCP services and private networking, plus functions and pipelines.
- [Heroku Scheduler alternative](https://inquir.org/heroku-scheduler-alternative): Full cron expressions and run history vs Heroku's fixed intervals.

## Serverless containers & runtimes

- [Serverless containers](https://inquir.org/serverless-containers): Container-backed Node.js, Python, Go functions — native modules declared in inquir.yaml, no edge isolate limits.
- [Docker serverless functions](https://inquir.org/docker-serverless-functions): Per-function Docker isolation with managed runtimes — no Dockerfile, fly.toml, or cluster YAML.
- [API key auth for serverless functions](https://inquir.org/api-key-auth-serverless-functions): Gateway-level auth before handler code runs — no auth boilerplate per function.
- [Deploy functions without Kubernetes](https://inquir.org/deploy-functions-without-kubernetes): Guide-style hub page.
- [Serverless REST API endpoints](https://inquir.org/use-cases/rest-api-endpoints): Public JSON APIs on functions — gateway auth, CORS, rate limits, and versioning without a routing monolith.

## Optional

- [Full LLM digest](https://inquir.org/llms-full.txt): Extended machine-readable facts and canonical links.
- [Sitemap (all public URLs)](https://inquir.org/sitemap.xml): Complete list of indexable marketing pages with EN and RU hreflang alternates.
- [Blog sitemap](https://inquir.org/sitemap-blog.xml): Published blog posts with hreflang alternates (not included in `sitemap.xml`).
- [Russian site (home)](https://inquir.org/ru/): Localized marketing and docs entry point.
- [About](https://inquir.org/about): Company and product background.
- [Pricing](https://inquir.org/pricing): Prices in US dollars on /pricing (another currency is converted by the payment agent at its rate on the charge date) and in roubles for Russian buyers on /ru/pricing; what each plan includes, delivery, cancellation and refunds.
- [Public offer](https://inquir.org/ru/offer): Contract for paid plans (Russian; English translation at /offer is informational).
- [Privacy policy](https://inquir.org/privacy): Personal data processing policy.
- [Terms of use](https://inquir.org/terms): Usage terms for all users, including the free plan.