# Deployment.io > Software built and deployed to your specification. Deployment.io combines its senior engineers, AI agents, and its software delivery platform to take software from specification to production in your cloud. Customers define the scope and acceptance criteria; Deployment.io owns delivery for managed engagements. ## What we build - **Build**: New features, services, and enterprise integrations. https://deployment.io/solutions/build/ - **Modernize**: Framework and dependency upgrades, migrations, and coordinated changes across repositories. https://deployment.io/solutions/dependency-modernization/ - **Remediate**: Implement and verify engineering changes behind security findings, compliance requirements, and enterprise commitments. https://deployment.io/solutions/security-compliance/ ## How delivery works Before any work begins, Deployment.io and the customer agree the scope, the acceptance criteria, and what is explicitly out of scope. That agreement is what a fixed price is quoted against. The delivery loop then runs on every piece of work: **Plan → Build → Verify → Deploy**. 1. **Plan**: A session reads the relevant repositories, services, and dependencies, and produces a specification the customer's team reviews before anything is written. 2. **Build**: Deployment.io’s senior engineers direct AI agents, which implement the specification in isolated environments. 3. **Verify**: Changes run against the customer's build, tests, and the acceptance criteria agreed in the plan. Work that fails required verification does not advance. 4. **Deploy**: Deploy into the customer's cloud after the designated reviewer approves production through the agreed release process. Coding agents provide tools for existing engineers. Traditional outsourcing provides engineering capacity. Deployment.io takes responsibility for delivering a defined project through verification and production deployment. ## Case studies Work on Deployment.io's own product, not customer engagements. Each figure is stated with the scope it was measured under. Full text: https://deployment.io/case-studies/ - Vite 4 to 8 upgrade (https://deployment.io/case-studies/vite-4-to-8-upgrade/): one Task produced the upgrade's pull request, merged on 27 August 2026. The bundler's production-build step fell from 33.49s to 4.29s (7.8x) in the recorded build environment, with 12% less JavaScript on the entry page's critical path. - Session security across three repositories (https://deployment.io/case-studies/session-security-across-three-repositories/): one Task changed a shared Go library, the API, and the dashboard, opening three pull requests 22 minutes after it started; that figure excludes specification, engineer review, production rollout, and a follow-up correction. Engineer review resolved 4 issues before merge. Deployed on 16 September 2026; a live-stream regression found afterwards was fixed in a follow-up release. - Security findings triage and remediation (https://deployment.io/case-studies/security-findings-triage-and-remediation/): from a synthetic nine-page assessment of our own code, 8 findings were checked against the code and 2 deliberately incorrect claims were rejected with file and line references. One selected remediation became two pull requests 7 minutes after its Task started, excluding assessment, specification, engineer review, and rollout, and was reviewed, merged, and deployed to production. Only that selected remediation is claimed as fixed. ## How we work and pricing Managed delivery includes Deployment.io’s senior engineers using the Deployment.io platform to coordinate the plan, build, verify, and deploy loop. The customer sets priorities, makes key decisions, and approves production. Engagements begin with an agreed scope, deliverables, acceptance criteria, timeline, payment terms, and fixed-price proposal. After an initial project, customers can continue on a monthly engagement: an agreed monthly fee and reserved delivery capacity for a prioritized stream of features, upgrades, and fixes, with each item scoped before work begins. See https://deployment.io/pricing/ for current engagement information and https://deployment.io/managed/ for responsibilities, onboarding, visibility, and handover. ## Platform Teams can also operate the Deployment.io platform themselves. It gives coding agents shared context, coordinated execution across repositories, verification, and deployment workflows. MCP tools connect coding agents to deployment and environment operations. See https://deployment.io/platform/ and https://deployment.io/docs/coding-agents/ for platform capabilities and setup. The platform supports static sites on S3 and CloudFront, web services on ECS, and development, staging, and production environments. ## Security and data handling - The runner checks out repositories, builds software, and executes agent tasks in the customer's cloud account with configured permissions. - Job state, logs, and metadata flow to the Deployment.io control plane to coordinate work and display progress. Runner-to-control-plane traffic uses TLS. - Agents send model requests to the configured AI provider. Running an agent in the customer's cloud does not mean every model request stays there; provider handling depends on the service and account configuration. - Provider API keys are encrypted at rest, masked in the interface, and supplied to agent containers at execution time. - Managed production releases require the designated customer's approval through the agreed release process. Platform approval policies are configurable per agent key. - Runner roles, team permissions, and provider credentials should be reviewed during setup and handover. See https://deployment.io/security/ for the security architecture. ## Key pages and documentation - Homepage: https://deployment.io/ - How we work: https://deployment.io/managed/ - Platform: https://deployment.io/platform/ - Use cases: https://deployment.io/solutions/ - Pricing: https://deployment.io/pricing/ - Security: https://deployment.io/security/ - Case studies (how we build and improve our own product with Deployment.io): https://deployment.io/case-studies/ - Contact: https://deployment.io/contact/ - Documentation: https://deployment.io/docs/ - Glossary: https://deployment.io/docs/introduction/glossary/ - Coding agents and MCP: https://deployment.io/docs/coding-agents/ - Runner architecture: https://deployment.io/docs/runner-installation/architecture/ - AI providers: https://deployment.io/docs/ai-providers/llm-providers/ - Team permissions: https://deployment.io/docs/role-based-access-controls/ - Guides: https://deployment.io/docs/guides/ - Blog: https://deployment.io/blog/ - Open-source skill: https://github.com/deployment-io/skills - Full documentation text: https://deployment.io/llms-full.txt ## Contact - General inquiries: hello@deployment.io - Security inquiries: security@deployment.io