--- title: "ShipDocs for engineering teams: what you gain" description: What dev teams gain when they adopt ShipDocs - onboarding velocity, always-current architecture, AI-assisted dev workflows grounded in your actual codebase, defensible security posture, and a roadmap that closes every gap that matters. date: 2026-04-29 author: Alex Sofroniev tags: [engineering teams, onboarding, documentation strategy, enterprise] image: /OG.png --- If you run engineering at a company with more than five developers, two facts about your codebase docs are almost certainly true: 1. They exist somewhere - a wiki, a `docs/` folder, scattered in Slack threads, or all of the above. 2. They're out of date enough that nobody fully trusts them. ShipDocs is built for the second fact. This post walks through exactly what teams gain when they adopt it, the v2 / v3 features on the roadmap, and how ShipDocs becomes the single docs surface for your engineering team. ## TL;DR - ShipDocs auto-generates structured docs from your repo, hosts them, and grounds AI chat in them - for any codebase, on any stack. - Onboarding drops from days to hours. AI assistants stop hallucinating. Architecture conversations get concrete. Bus factor drops. - v2 ships team members + roles, versioned docs, scheduled / on-push regeneration, multi-language docs, and a public docs portal with templates. - v3 ships SSO / SAML, Slack / Teams integration, and a visual dependency diagram. - Paid plans start at $29/mo (Pro) and $29/mo (Advanced), with a 3-day free trial. ## What you gain ### 1. Onboarding velocity The most measurable win. New engineers spend their first week or two pattern-matching against the codebase. Without good docs that's a multi-day cost; with structured per-component docs, it collapses to hours. A few things that shift: - New hires can read a complete Overview before they ever clone the repo. - Per-component docs let them go deep on the area they'll be owning, with explicit dependencies and exports. - Their AI assistant (Cursor, Claude Code, Continue, Cody) gives substantially better answers because the docs are part of its context. - The senior engineers who would otherwise have been the onboarding bottleneck get their time back. If a single hire saves three days of senior-engineer time, ShipDocs has paid for the year. ### 2. Always-current architecture map Internal architecture docs in Confluence have a lifecycle: someone writes them under pressure, the team uses them for two months, the architecture changes, the docs don't, and within a year nobody links to them. Auto-generation breaks that cycle. ShipDocs generates from the actual current state of the repo. The docs reflect what's deployed, not what someone meant to write down 18 months ago. This shows up especially in: - Service maps in microservice orgs. - "Where does X live" questions across a monorepo. - "What does this worker actually depend on" - answered from the manifest, not from memory. ### 3. AI-assisted dev across the team Every member of your team is using AI assistants whether you have a policy on it or not. Their effectiveness varies wildly based on context. ShipDocs gives the whole team a shared, grounded context layer: - Internal AI chat (over your projects' generated docs) gives consistent answers regardless of who's asking. - Engineers using external AI tools can paste the relevant doc into the prompt for grounded answers. - The Markdown twins served at every doc URL are LLM-friendly, so retrieval-style integrations work without scraping HTML. The team-wide effect: junior engineers ask better questions, senior engineers get fewer interruptions, and AI-generated code is meaningfully more correct because the model isn't guessing about your codebase. ### 4. Lower bus factor When the engineer who built the auth layer leaves, the cost is everyone else having to rebuild a mental model from source. With per-component docs that capture exports, dependencies, and the place each piece occupies in the architecture, the cost drops materially. It doesn't replace the tribal knowledge of *why* a decision was made - that still belongs in commit messages and design docs - but it captures *what's there*, which is the larger fraction of what departing engineers carry. ### 5. Defensible security posture Most internal docs tools store your repo content in ways that aren't easy to audit: - AES-256-GCM at rest with **per-project encryption keys**, wrapped by a master key in a KMS-backed vault. Each project has its own DEK; compromise of one doesn't compromise the others. - **Your code is never used to train AI models.** Not by us, not by the model providers we use. We pay providers directly on zero-retention API endpoints. Our [robots.txt](https://shipdocs.sh/robots.txt) advertises `ai-train=no` to every crawler. - **Soft-delete grace window of 90 days** for projects, chats, and accounts. After that, cron purges hard. - **Hardened transport.** TLS in transit, isolated processing environments, audit-logged admin reads. If you've been holding back from cloud docs tools for security reasons, the per-project keys plus the zero-training guarantee are designed for exactly that case. ## What's already shipping vs what's coming Some teams ask "is ShipDocs ready for us today?" The honest answer: yes for codebase docs, AI-grounded chat, encryption, and showcase publishing. Versioning, multi-seat, and the polished public portal are landing in the next two releases. | Capability | Status | |-------------------------------------------|-----------------------------------------------| | Auto-generated per-component docs | Shipping today | | AI chat grounded in your codebase | Shipping today | | WYSIWYG Markdown editor | Shipping today | | Per-project encryption + no AI training | Shipping today | | Public showcase pages | Shipping today | | Multi-source import (Git / zip / CLI) | Shipping today | | Markdown twins for AI agents | Shipping today | | Team members + roles | **v2 - coming soon** | | Scheduled / on-push doc regeneration | **v2 - coming soon** | | Doc versioning tied to git refs | **v2 - coming soon** | | Public docs portal with templates | **v2 - coming soon** | | Multi-language docs via DeepL | **v2 - coming soon** | | Simplified docs for non-technical readers | **v2 - coming soon** | | Visual dependency diagram | **v3 - on the horizon** | | Slack & Microsoft Teams integration | **v3 - on the horizon** | | SSO / SAML | **v3 - on the horizon** | That roadmap closes every gap most teams ask us about. ## How ShipDocs slots in alongside what you already have ShipDocs is the codebase-doc layer. Whatever your team uses today for narrative product docs, runbooks, and decision records keeps working - ShipDocs sits next to it as the place codebase truth lives. As v2 ships the public docs portal, more teams move customer-facing surfaces into ShipDocs too; you can do that on your own timeline. Concretely, on day one ShipDocs is the right home for: - Per-service / per-module docs that have to stay current. - Internal architecture maps and Overview docs for onboarding. - Grounded AI chat for "what does this service do" / "where is X handled" questions. - Public docs URLs for OSS projects via showcase pages. - Sensitive codebase docs (private repos, encrypted at rest, never used for AI training). ## How to roll it out Three-stage rollout works well: **Stage 1 - pilot (one week).** One staff engineer connects two or three of the most-asked-about services. They generate docs, share the links internally, and answer "is this useful for the team?" with data. Low cost, no commitment. **Stage 2 - team adoption (two to four weeks).** Connect every service the team owns. Each engineer flips on the chat and runs it for a sprint. Track: how many "where does X live" Slack questions get answered with a doc link instead of a senior-engineer interruption. **Stage 3 - org-wide (gradual).** Expand to other teams. Pair with your internal AI assistant if you have one - ShipDocs' Markdown twins are designed for retrieval. Set scheduled regeneration once that's live in v2. You don't need to migrate any existing docs in or out. ShipDocs sits next to what you have. ## Pricing for teams | Plan | Price | Team fit | |-----------------|------------|-----------------------------------------------------------| | Pro | $29/mo | Focused projects — 5 repos, 500k chat tokens/mo | | Advanced | $29/mo | One engineer or small team, up to 20 repos, unlimited chat| | Team (v2) | Per-seat | Multi-seat with role-based access | | Team + SSO (v3) | Per-seat | SSO/SAML, audit logs, enterprise controls | If you need multi-seat or SSO sooner than v2, get in touch and we'll set you up. ## Try it with your team Pick one service or one repo. Connect it. Generate. Share the link in your team channel. The pilot is reversible: cancel anytime, soft-delete is 90 days, hard-delete after. - **Get started →** [shipdocs.sh](https://shipdocs.sh/#hero) - [Showcase](https://shipdocs.sh/showcase) - see ShipDocs output on real OSS projects before deciding. - [Privacy & encryption](https://shipdocs.sh/privacy) - full policy. ## FAQ **Is ShipDocs SOC 2 / ISO 27001 compliant?** Audit work is in progress. Architecture is built to the controls (encryption at rest, audit logging on admin reads, GDPR-aligned retention) but the certification process takes time. If you need a current attestation for procurement, get in touch. **How does ShipDocs handle private repos?** Connect via OAuth (GitHub, GitLab, Bitbucket [coming soon]) - same scopes as any read-only docs tool. Repos are pulled, processed, and stored encrypted with AES-256-GCM and a per-project key. Your code is never used to train AI models. **Will my source code be used to train AI models?** No. Not by us, not by the model providers we use. We pay model providers directly on zero-retention API endpoints, so prompts and completions are never retained or used for training. **Where does ShipDocs sit if we already have a docs site?** ShipDocs becomes the codebase-doc layer immediately and the polished public portal as v2 lands. Most teams adopt for the codebase surface first and migrate the public-facing surface in once the v2 portal ships. See [the comparison post](https://shipdocs.sh/blog/shipdocs-vs-mintlify-gitbook-notebooklm-cursor-sourcegraph) for the full breakdown. **What's the team plan?** Advanced tier today (up to 20 repos, unlimited chat). Multi-seat with role-based access ships in v2; SSO/SAML and audit logs ship in v3. If you need multi-seat sooner than v2, contact us. **How do we evaluate without signing up?** Browse [the public showcase](https://shipdocs.sh/showcase) - full ShipDocs runs on popular OSS projects, no signup needed. That gives you a real sense of output quality. --- **Ready to start the pilot?** [Get started →](https://shipdocs.sh/#hero)