Skip to main content
All integrations
AssessmentsOn the roadmap

CoderPad and Expert Hire

Bridge live coding sessions in CoderPad with Expert Hire evaluation evidence for unified review.

Centralize coding-session signal alongside structured AI evaluation to give reviewers one place to look.

Nothing on this page describes shipped behaviour. It describes the mapping we are building to.

CoderPad connected to the Expert Hire workflow
A practical path from Expert Hire to CoderPad.
Workflow coverage

What the CoderPad connector is being built to do.

These are the workflows this connector is scoped around. They describe the target, not shipped behaviour. Until it lands, the same outcomes are reachable through the MCP connector and the built-in pipeline board.

01

Attach CoderPad session links and outcomes to Expert Hire records.

02

Trigger CoderPad sessions when candidates progress past async screening.

03

Surface combined evidence for reviewer decisions.

Whatever the connector moves, the evaluation does not change. Every stage is one of five modes: resume screening, AI interview, human-led interview, AI prompt assessment or coding test. All of them score onto one rubric, and every score traces back to the transcript, the live code, the skills you weighted and the JD you set up.

Planned field mapping

What moves, and which way.

Expert Hire already runs its own sandboxed coding test with real execution against hidden test cases. An assessment connection exists so external signal does not end up living in a second tool.

This is the mapping we are building the CoderPad connector to. It is a plan, not a description of live behaviour.

  • Candidate

    Both directions

    Matched to one Expert Hire candidate so signal from both systems lands on a single profile.

  • External assessment result

    Into Expert Hire

    The score from the other system, held next to the built-in coding test result rather than replacing it.

  • Combined reportcard

    Out of Expert Hire

    One reviewer-facing report with the skill breakdown, the code and the transcript, instead of two numbers in two places.

What you are looking at: the built-in pipeline board every plan gets, with live counts on each stage column, source tags on each card and search by name or email. The later stages appear once the frame is wide enough for them. This is the side of the mapping Expert Hire owns. Sample data.

The board is the fallback, not the consolation prize.

Nothing on this page has to ship before you can hire. The pipeline board above is included on every plan, it is where candidates land whether they arrive by upload, referral or an assistant over MCP, and it is where reportcards attach. A connector changes where the record lives, not whether you can run a round.

Agencies get the same thing per client. One workspace runs many client companies, each with its own roles, pipeline and rubric, and data never pools across clients.

Auth, keys and plan access

How the connection is secured, and what it costs to open.

Every integration on the platform authenticates through the same surface, whether it is a packaged connector or something your team builds. Here is that surface, and the plan tier this particular one sits behind.

OAuth on the MCP server.

The remote MCP server that exposes Mia to AI assistants is OAuth-protected. Financial actions, team-admin changes and destructive operations are excluded by design.

Audit trail on every action.

Role-based access control, audit logs, encryption in transit and at rest, and single-tenant separation with no shared candidate pool. Export any time.

Plan gating

No plan gate has been published for this category. Coding tests cost three Hire Credits and are available on every plan; the desktop proctoring app adds no credits and unlocks at Growth.

Using Expert Hire as your own ATS is included on every plan, including the free trial. Team members are unlimited on every tier, so nothing here is priced per seat.

The MCP server is live at mcp.experthire.cloud and connects over OAuth.

Rollout plan when it ships

Four steps, in this order.

This is how a CoderPad rollout would run once the connector lands. Design partners who register interest get walked through it first.

  1. 1Connect CoderPad workspace
  2. 2Define session triggers
  3. 3Map reviewer ownership
  4. 4Pilot per role family
FAQ

CoderPad questions, answered.

Ship state, field mapping, plan gating and what to do in the meantime.

Still not sure it fits your stack? Tell us what you run and we will answer honestly.

CoderPad is not live yet. You can still start today.

Register interest and we will tell you honestly where this connector sits. In the meantime the built-in pipeline board is on every plan, and the MCP server and the MCP connector already reach most of what this one would do.

30-day free trial with 75 Hire Credits. Card required, nothing charged for 30 days.