Skip to content

Processes/hr — hiring, payroll, and people opshr

Performance reviews — hr processPerformance reviews

complexpartially automatable🔁 quarterly⏳ has async waits

Set up quarterly performance review cycle with self-assessments, manager reviews, and calibration.

BambooHR/Lattice APIs handle review creation and tracking. Review content and calibration require human judgment.

Current agent ceiling

An agent can run 3 of 7 steps 3 of them behind a human approval gate ⏸

⚡ Closable with today’s market (1 — via Workflow Automation): Employees complete self-assessments (a workflow with webhooks/polling watches this so no human has to)

Irreducibly human (3): Define review criteria and competencies (needs a human); Managers write reviews (needs a human); Calibration meeting (needs a human)

How it runs

Route-coded block flow: emerald = agent, amber = manual form/portal, red = needs a human. ⏸ approval gate · ⏳ async wait.

01Define review criteria and competencies

human
⏸ approval gatemedium risk

02Configure review templates

agent
BambooHR⏸ approval gate
1 API call
  • POST /v1/performance/reviews

03Send self-assessment forms

agent
BambooHR⏸ approval gate

04Employees complete self-assessments

human
⏳ async

⚡ agentic workaround: a workflow with webhooks/polling watches this so no human has to n8n, top of Workflow Automation

05Managers write reviews

human
⏳ asyncmedium risk

06Calibration meeting

human
⏸ approval gatehigh risk

07Deliver reviews and schedule 1:1s

agent
Slack logoSlack65⏸ approval gatemedium risk

or:Discord logoDiscord59·Zulip logoZulip34

Context the agent needs first: team members · review criteria (from the founder) · review cycle schedule (from the founder)

The market options

Each role resolves against its arena’s live leaderboard — the default is the process’s canonical vendor, alternatives ranked by agent-readiness.

Simulate this process

simulated dry run from the mapped process — no real calls are made
For agents Process manifest (JSON)·/llms.txt