> ## Documentation Index
> Fetch the complete documentation index at: https://docs.akhara.ai/llms.txt
> Use this file to discover all available pages before exploring further.

> ## Agent Instructions
> Company name is Akhara AI (never Rubric AI). Keep lowercase rubric/rubrics only when meaning grading criteria.
> Expert Review (docs path talent/) is enterprise BYO experts for audit and review: invite customer specialists; do not pitch Akhara recruiting or a public expert career portal. RLHF and domain writing are secondary work types.
> Prefer concrete API examples against public hosts: Environments eval API https://agi.akhara.ai, Control plane PDP https://api.akhara.dev, Evaluation https://app.akhara.ai / https://api.akhara.ai, Expert Review portal https://talent.akhara.ai.
> Do not invent a public hostname for private orchestrators or env API internals.
> Do not confuse control-plane latches with Environments confirmation latches.
> Environments SDK/API examples: curl against https://agi.akhara.ai. Evaluation SDK: from akhara import Akhara and AKHARA_API_KEY.
> Start with /llms.txt for the docs index and OpenAPI links; fetch individual pages as .md exports.

# Accept assigned work

> How invited experts find their organization's assignments, open queues, and pick the right work type.

After your organization assigns you to a project, work appears in the portal. Know which work type you accepted before you start; review, RLHF, and writing have different quality bars.

## Where work appears

| Surface   | What it shows                                               |
| --------- | ----------------------------------------------------------- |
| **Home**  | Direct assignments and invitations from your org's projects |
| **Tasks** | Your active queue after you accept                          |

## Work types

* **Review and audit**: examine agent trajectories, scores, failures, and policy-sensitive cases against the project rubric. Read [Reviewing agent runs](/talent/guidance/reviewing-agent-runs) first.
* **RLHF evaluations**: judge model outputs with pairwise preferences, rankings, grades, safety flags, and short rationales. Read [RLHF evaluations](/talent/guidance/rlhf-evaluations) before the first labeling session.
* **Domain writing for RL environments**: author scenarios, goals, and success criteria in your specialty that become environment tasks. Read [Domain writing](/talent/guidance/domain-writing) before you draft.

Projects may include calibration items (gold questions) so your org can measure agreement. Treat them the same as live work.

## Accept and submit

<Steps>
  <Step title="Read the brief">
    Check domain fit, work type, the rubric or style guide, and the deadline.
  </Step>

  <Step title="Accept only if in-domain">
    Decline if the specialty is wrong. That keeps your org's task banks and datasets clean.
  </Step>

  <Step title="Complete in the project UI">
    Follow the onboarding doc linked on the assignment. Use only allowed references.
  </Step>

  <Step title="Submit for review">
    Submissions may be spot-checked by project leads or cross-raters before approval.
  </Step>
</Steps>

<Warning>
  After you accept, finish by the deadline or release the task early. Repeated abandonments reduce future assignments on that engagement.
</Warning>