MCP

AI-agent format

Assemble a new-hire ramp kit from curated videos

Point new hires at the library, generate study notes per module, and let them ask the library questions instead of you.

Engineering managerTeam Knowledge & Library Opslibrary_listdeep_contextrun_youtube_tool
The use case

The problem

Every new hire gets the same onboarding artifact: a doc titled "Useful Links" containing forty videos nobody has watched in order, assembled by the last person who onboarded and never pruned. The hire dutifully opens three, drowns, and switches to interrupting their manager — which is the actual onboarding system at most companies: senior time, retail-priced.

The kit everyone means to build — curated material, in sequence, with notes, and someone to answer questions — fails on maintenance cost. Every piece of it is now generatable from a curated library: the curation stays human (that's the manager's taste), everything downstream stops costing anyone's week.

The mechanics

How veridive MCP helps

library_list turns your curated imports into a syllabus: the onboarding-tagged videos, listed and ordered — the ramp path, visible instead of tribal. run_youtube_tool's study-notes generates the companion notes per module, so hires read before they watch and retain after.

deep_context is the piece no links-doc ever had: the hire asks the library their questions — "how does our deployment flow work?" — and gets answers from the curated material, cited to the exact minute, before spending a senior engineer's afternoon. The manager's interruption load drops to the questions that genuinely need a human, which is what senior time was supposed to be for.

Step by step

How to run it

  1. 1
    Connect veridive MCP

    One-time setup: add the veridive MCP server to Claude, ChatGPT, Cursor or any MCP client and generate your API key. Takes about two minutes.

  2. 2
    Curate once, tag as the ramp

    The videos that actually onboard people — internal talks, key conference sessions, tool deep-dives — imported and tagged 'onboarding'.

    List everything in my library tagged 'onboarding' — that's the ramp path. Order it from fundamentals to specifics.
  3. 3
    Generate the module notes

    Study notes per video: the hire's pre-read and post-watch reference.

    Generate study notes for each video in the onboarding tag — structured, with the key concepts up front.
  4. 4
    Hand over the question channel

    The hire's assistant queries the same library — cited answers first, human escalation second.

    From the onboarding library: how does <team process/tool> work? Cite the videos and moments so I can rewatch.
Before vs after

Problems it solves

  • Onboarding runs on a stale links-doc plus manager interruptions — senior time at retail price.
  • Curated ramp kits die of maintenance cost: notes, sequencing and Q&A all cost someone's week.
  • Hires can't distinguish the critical video from the forty optional ones.
  • The same questions get answered live for every hire, cited to nobody's memory in particular.
The payoff

Why it's valuable

  • The manager's cost drops to curation — taste stays human, production goes automatic.
  • Hires get a sequenced syllabus with notes instead of a link dump.
  • The library answers the recurring questions with citations; humans keep the judgment questions.
  • Every hire improves the kit: their unanswered questions flag the library's next import.
Copy & paste

Example prompts

List my 'onboarding'-tagged library, ordered fundamentals-first.
Study notes per onboarding video, key concepts up front.
From the onboarding library: how does <process> work? Cited moments.
Questions

FAQ

What goes in the ramp library?

Whatever you already send hires: recorded internal talks on YouTube, the conference sessions that explain your stack's philosophy, tool deep-dives, your own team's public tech talks. Curation is the manager's judgment — that's the human part that stays.

How do hires access it?

The MCP connection is shareable config — the same setup the team uses (see the research-library use case). The hire's own assistant queries the shared library from day one.

Does this work for non-engineering onboarding?

Anywhere the ramp material is video: sales methodology talks, product training, industry orientation (pair with the weekend-onboarding use case for the industry layer).

Keep going

Related use cases

Stand up a research library your whole team queries

Curate once, query forever: shared imports become a common brain with citations instead of link-dump Slack threads.

Attach quizzes to your training library

Generate a quiz per training video and make completion mean something — comprehension, not checkbox.

Onboard into a new industry over a weekend

Ask the naive questions at scale — grounded answers with citations build a mental map faster than any reading list.

Run this in your own assistant

Connect the veridive MCP server to Claude, ChatGPT or Cursor and paste the prompts above — setup takes about two minutes.

Get set up Read the docs