Onboarding

Make "how we do things here" searchable from day one.

New hires take 8-12 months to become fully productive. Most of that time is spent learning unwritten rules and tribal knowledge.

The problem

The onboarding doc is a 50-page Google Doc that's always outdated. Real knowledge lives in Slack threads, meeting recordings, and the heads of people who've been around longest. New hires learn by interrupting busy colleagues.

How it works

Here's how you build your knowledge graph step by step using the ai CLI.

Add team structure and responsibilities

Help new hires understand who does what and who to go to for what.

ai add "Platform team owns infrastructure, CI/CD, and \
  developer experience. Product team owns features, \
  A/B tests, and user-facing APIs. Data team owns \
  pipelines, analytics, and ML models. On-call \
  rotation is shared across Platform and Product." \
  --title "Team Structure" -t note -d operations

Add "how we work" processes

The unwritten rules that take months to absorb — make them explicit and searchable.

ai add "Sprint cadence: 2 weeks, starts Monday. \
  Standup: async in #eng-standup by 10am. \
  PR reviews: 24h SLA, minimum 1 approval. \
  RFC process: write doc, share in #eng-rfcs, \
  3-day comment period, then decision meeting. \
  Demo day: every other Friday at 3pm." \
  --title "How We Work" -t process -d operations

Add tool setup guides

Document the tools and access new hires need, so they can self-serve instead of waiting on IT tickets.

ai add "Day 1 setup: 1. Clone mono-repo, run 'pnpm install'. \
  2. Copy .env.example to .env.local, get secrets from \
  1Password 'Engineering' vault. 3. Run 'docker compose up' \
  for local Postgres + Redis. 4. Request access: AWS \
  (ask #platform), Datadog (#observability), \
  Linear (#eng-ops)." \
  --title "Dev Environment Setup" -t process -d technology

Add tribal knowledge and FAQs

The questions every new hire asks in their first month — answer them once, permanently.

ai add "Why is the users table called 'accounts'? Legacy \
  naming from v1 when we had multi-user accounts. \
  Why do we use Cloud Run instead of ECS? Decision \
  made in Q3 2024 to reduce ops overhead. \
  Why two payment providers? Stripe is primary, \
  Adyen is fallback for EU regulatory compliance." \
  --title "Engineering FAQ" -t note -d technology

Add internal terminology

Every company has its own language. Define it so new hires don't spend weeks decoding acronyms.

ai add "POP = Payment Orchestration Platform (our product). \
  TRX = transaction record in the database. \
  Golden path = the happy-path user flow we optimize for. \
  War room = incident response channel, auto-created \
  by PagerDuty. LGTM = 'looks good to me' (PR approval)." \
  --title "Internal Glossary" -t note -d operations

New hire searches from day one

Instead of interrupting colleagues, new team members search the knowledge graph.

ai search "how do I set up my development environment"
ai context "what is the sprint process and who is on what team"

Returns the setup guide, team structure, and sprint process — all connected and current.

Onboarding is now self-service

New hires have a searchable knowledge base from their first day. Tribal knowledge is captured, connected, and always up to date.

New hires ramp up in weeks instead of months

Fewer interruptions for senior team members

Tribal knowledge survives employee turnover

AI tools give new hires instant answers grounded in your actual processes

Start building your knowledge graph

Free during beta. No credit card required.