Optimize onboarding mentorship portfolio

What it's for

Gives engineering leaders an evidence-backed mentoring plan that balances faster autonomy against scarce senior attention, timezone fit, launch risk and financial downside instead of assigning the nearest available senior engineer.

What you give it

Every tool is a pure function: it computes from what you pass it and reads nothing on its own. Inputs split into evidence you supply from your connected systems, calibration your team owns, and numerical controls that affect precision but never the result's meaning.

{
  "budget": 1000,
  "causal_studies": [
    {
      "assignment_design": "cluster_randomized",
      "control_count": 30,
      "design_reliability": 0.9,
      "id": "paired-review-study",
      "interference_design": "cluster_accounted",
      "log_time_ratio_effect": -0.35,
      "option_type": "paired_review",
      "prospectively_registered": true,
      "standard_error": 0.05,
      "transport_weight": 0.9,
      "treated_count": 30
    }
  ],
  "mentor_capacity_hours": {
    "mentor-0": 8,
    "mentor-1": 8
  },
  "mentorship_options": [
    {
      "cost": 0,
      "dependency_option_ids": [],
      "excluded_option_ids": [],
      "id": "baseline-0",
      "is_baseline": true,
      "knowledge_fit": 1,
      "learner_hours": 0,
      "mentor_hours": 0,
      "mentor_ref": null,
      "mentor_response_reliability": 1,
      "option_type": "baseline",
      "timezone_overlap": 1,
      "unit_id": "unit-0"
    },
    {
      "cost": 100,
      "dependency_option_ids": [],
      "excluded_option_ids": [],
      "id": "pair-0",
      "is_baseline": false,
      "knowledge_fit": 0.95,

What you get back

This is the actual output of running the example above — computed by the same function the platform calls, not an illustration.

{
  "decision": "mentorship_portfolio_selected",
  "evidence": {
    "admissible_option_type_count": 1,
    "rejected_option_counts": {},
    "rejected_study_counts": {}
  },
  "gates": {
    "budget": 1000,
    "maximum_expected_autonomy_days": null,
    "maximum_loss_cvar": null,
    "minimum_knowledge_fit": 0,
    "minimum_on_time_probability": 0,
    "minimum_timezone_overlap": 0
  },
  "interpretation": "Effects are admitted only from prospective controlled designs, shrunk toward no effect by design reliability and transported conservatively. Mentor capacity, response reliability, full cost, shared loss, and interference are explicit. Opaque assignments require accountable human review and cannot rank, reward, penalize, hire, or fire anyone.",
  "method": "causal_interference_aware_mentorship_portfolio_search_v1",
  "pareto_frontier": [
    {
      "all_units_on_time_probability": 0.8,
      "expected_loss": 820,
      "loss_cvar": 4100,
      "selected_option_ids": [
        "baseline-0",
        "baseline-1"
      ],
      "total_cost": 0
    },
    {
      "all_units_on_time_probability": 0.8,
      "expected_loss": 602,
      "loss_cvar": 2610,
      "selected_option_ids": [
        "baseline-0",
        "pair-1"
      ],
      "total_cost": 100
    },
    {
      "all_units_on_time_probability": 1,
      "expected_loss": 204,
      "loss_cvar": 220,
      "selected_option_ids": [
        "pair-0",

How it works

Forecasting & survival — Estimate when something completes or fails, with censoring and unresolved work handled honestly rather than dropped.

1. 1 Admit only prospectively registered randomized, cluster-randomized or frozen-propensity studies whose peer interference is absent or explicitly cluster-accounted; reject weak arms and shrink transported effects toward no effect by design reliability.
2. 2 Evaluate one baseline or mentoring policy per onboarding unit under coherent common scenarios, attenuate benefit by mentor response reliability while retaining full cost, enforce knowledge fit, timezone overlap, mentor capacity, dependencies, exclusions and mandatory support.
3. 3 Count direct lateness for every unit and shared launch/coordination loss once per group and scenario; enforce budget, expected-autonomy, on-time and CVaR gates, then return the exact or beam-search choice plus the non-dominated cost-loss-tail frontier.

Before you trust it

Every tool in the catalog ships with the conditions under which its answer is meaningful — and the conditions under which it should abstain instead of guessing.

Calibrating it to your org

Calibration workflow

1. 1 Define the management decision, target outcome, aggregate unit, privacy boundary, cadence, and prediction/intervention horizon for this organization.
2. 2 Build a tenant-scoped historical cohort using only information available before each prediction or decision; preserve zero periods, censoring, assignment probabilities, and unresolved outcomes when the method requires them.
3. 3 Estimate statistical parameters on training history, but obtain costs, utilities, risk tolerance, practical-effect thresholds, capacity, and policy constraints from accountable decision owners.
4. 4 Validate on later time windows or held-out aggregate units at the deployment grain, against a simple baseline and the function-specific validation strategy.
5. 5 Deploy only if the returned decision clears evidence, overlap, calibration, robustness, and guardrail checks; warning, unsupported, schema-gap, and fallback decisions are abstentions.
6. 6 Monitor realized outcomes, data drift, coverage, and decision regret; recalibrate at a governed cadence or after a detected regime/definition change, never merely because a stakeholder dislikes the result.

Call it from your AI

You don't wire up 388 tools in your MCP client. The GitRevio MCP server exposes 20 tools, three of which let an agent search the catalog, read a tool's schema, and run it — so the assistant finds this one on its own.

gitrevio_capabilities_search
  { "q": "choose an interferenceaware mentorship portfolio using" }
  → finds "optimize_onboarding_mentorship_portfolio"

gitrevio_capability_describe
  { "capability_id": "optimize_onboarding_mentorship_portfolio" }
  → returns the input schema and agent guidance shown on this page

gitrevio_capability_run
  { "capability_id": "optimize_onboarding_mentorship_portfolio", "arguments": { ... } }
  → returns the result shown above

Works in Claude Desktop, Claude Code, Cursor, Cline, Continue.dev, Goose and Aider. See the MCP server.

Related tools

See every tool in People, retention & knowledge →