liveBook · Manning

Team design

As explained in Delivering Machine Learning Projects: From design to deployment

Team design is the structuring of a project team’s roles, responsibilities, and composition to support delivery.

Team design turns an ML project roadmap into a resourcing plan: who can deliver each task, when they are needed, and how long their work may take. Team time is usually the largest project cost, so the design is central to a credible estimate. The plan is an envelope of resources and cost, not a promise that staffing will remain unchanged. section 3.2.2 section 3.2

Build the plan task by task:

  • Break the roadmap into tasks and assign each one a resource suited to its anticipated scale and difficulty. section 3.2.2
  • Estimate the kind of person and time each task requires; work through the estimates with pre-project colleagues or senior delivery staff. section 3.2.2
  • Account for data, modelling, and domain-alignment challenges when deciding which specialist capabilities are needed. Business translators, for example, bridge the domain and ML specialists. section 3.2.2
  • Create and socialize the design alongside the project estimate, then review the pre-project work against a checklist. section 3.2

Finalize the design and assignments in sprint 0. Confirm that reserved people can actually join and begin onboarding them. If a key person is unavailable, find a replacement or redesign the team; record the disruption in the risk register and escalate inadequate resourcing. If the original staffing cannot be secured, replan when work is done or consider external contractors, recognizing their potential cost to the commercial case. section 4.2

How Delivering Machine Learning Projects: From design to deployment covers it

How deep it goes
Works through task-level allocation and time estimates, then addresses unavailable staff and mitigation; does not prescribe fixed staffing ratios.
Examples
Uses team-design tickets for pre-project planning and sprint 0, plus a staffing disruption involving unavailable team members.
What it assumes
Assumes a project roadmap and outline of the work are available before resources are allocated.
What you can do after
Practice/perform task-by-task resourcing, confirm assignments, and respond when the planned team cannot be assembled.
How it connects
Links team design to project roadmaps, cost estimation, specialist roles, sprint planning, and risk registers.

This explainer was generated by Manning's knowledge graph from the text of Delivering Machine Learning Projects: From design to deployment. The section links point to the parts of the book it draws on.