Plugins

Design embedding pipelines — model selection, batching, normalization, index refresh strategy.

Audit embedding infrastructure — model drift, index freshness, query latency, coverage gaps.

Optimize similarity search — ANN index tuning, hybrid search, reranking, query expansion.

Experiment Design Engineer — Experiment design — A/B testing, statistical power, experiment tracking, causal inference

Analyze A/B test results

Design an A/B test

Design eval harnesses — task schemas, metrics, dataset versioning, eval-as-code patterns.

Audit existing experimentation infrastructure

Build automated regression suites — golden sets, threshold alerting, CI integration for model changes.

AI Operations Team — Evals: Eval harness design, benchmark suites, automated regression, human eval orchestration.

Feature Engineer — Feature engineering — transformations, encodings, feature stores, pipeline design

Design a feature engineering pipeline for a ML problem

Audit feature engineering code for leakage and quality issues

Design or audit a feature store

Infrastructure Specialist Team — Finop: Cloud cost optimization — FinOps practices, rightsizing, reservation strategy, cost attribution

Finop skill: finop-audit

Finop skill: finop-recon

Finop skill: finop-reserve

Model Training Engineer — Model training — algorithm selection, hyperparameter tuning, training infrastructure

Audit existing model training code

Design a model training pipeline

Design a hyperparameter tuning strategy

People engineer — org design, hiring pipelines, compensation frameworks, onboarding playbooks, performance management, and human-to-agent migration

Corporate Governance Advisor — board resolutions, cap table, equity plan docs

Draft board resolutions and corporate consent documents

Draft equity plan documents and option grant agreements

Survey corporate governance artifacts and flag gaps

API Quality Gate Engineer — API quality gates — linting, style enforcement, breaking change CI, and API governance

Integrate API quality gates into CI

Design an API linting ruleset with style rules and severity levels