> ## Documentation Index
> Fetch the complete documentation index at: https://docs.heymilo.ai/llms.txt
> Use this file to discover all available pages before exploring further.
> View, download, and share candidate reports (score reports and PDFs). Review candidates from the interviewer list, check interview URLs, and run actions.
# Reviewing Candidates
### 1. Open an interviewer (agent)
You’ll see the interviewer’s header: job title and ID (e.g. `627-seasonal-sales-associate`), status (Active), timeline (created, active for X more days), budget/spend, subscribers, and **Share** / **Actions**.
Tabs under the header: **Candidates**, **Sally** (or agent name), **Settings**, **Integrations**, **Analytics**. Stay on **Candidates** to review.
### 2. The candidate list
The list shows everyone in that interviewer’s pipeline:
* **Profile**: Name, email, and (if shown) location/flag.
* **Status**: Per stage: e.g. **Resume** (score or progress like 74% • 2/3), **Video** (score like 80% or “Video” if completed), or pending.
* **Last Engaged**: When the candidate last interacted.
Use **Search** to find by name or email. Use **Filter** and **Columns** to narrow by status, stage, or custom columns. **All Candidates (X)** lets you switch views (e.g. all vs completed).
### 3. Actions (list or after selecting candidates)
Click **Actions** to run bulk or single actions:
| Action | Use |
| ------------------------------------- | ----------------------------------------------------------------------------------------------------------------- |
| **Shortlist** | Move candidate(s) forward — they’re in. |
| **Dismiss** | Reject or remove from consideration. |
| **Nudge** | Send a reminder (e.g. to complete a step). |
| **Blast** | Send a bulk message to selected candidates. |
| Sync/Refresh | Refresh this candidate's profile so the latest data, statuses, and evaluations are reflected. |
| **Send Results to ATS** | Push scores and notes to your ATS. See [Integrations](../../integrations/ats-quickstarts) (ATS sync & writeback). |
| **Enrich** | Add or refresh data on the candidate profile. |
| **Create Cohort** / **Add to Cohort** | Group candidates for a role or talent pool. |
### 4. Candidate reports
After candidates finish their interviews, use the options below to **view**, **download**, or **share the candidate report** (also called the score report or PDF report).
HeyMilo uses **two different Download menus**: on the **Candidates list** and on a **single candidate profile**.
#### All candidates (list)
On **Candidates**, click **Download**:
* **Download Report**: General report.
* **Package Zip**: Opens **Download Candidate Files Zip**. Choose what to include (e.g. **Include PDF Report**, resume, video, audio). For the PDF: set **Report style** (**Modern** for multilingual, **Classic** for English only) and **Language**, and toggles such as chips, highlights, lowlights, transcript when shown. The zip is **emailed** to you.
* **Export CSV**: Candidates and scores (spreadsheet).
* **Export JSON**: For integrations or tools.
* **Export Candidate Feedback CSV**: Feedback / evaluation export.
**Bulk:** Select multiple rows, then **Download** — same style of options; delivery is usually **email** (ZIP).
#### One candidate (profile)
Open a candidate, then click **Download**:
* **View Report**: Opens the report in the **browser** (preview).
* **Download PDF**: Saves a PDF of the report. Set **Report version** and **Language** as needed; adjust chips, highlights, lowlights, and transcript. **Transcript** stays in the **original interview language** (not translated).
* **Package Zip**: Same **Download Candidate Files Zip** flow: optional PDF (same options as **Download PDF**), plus resume, video, audio; **emailed** ZIP.
**Note:** Multilingual PDFs use **Modern**; **Classic** is English-only.
More detail: [Scorecards & Analysis — Downloading & translating reports](https://docs.admin.heymilo.ai/wILbmLp94QU0dFy3Z6wB/reviewing-candidate-data/dashboard/scorecards#downloading-and-translating-reports). In-repo: Multilingual Support. API: **Export Interviews** in Public API documentation.
### 5. Click into a candidate
Click a row to open that candidate’s full review.
**Candidate card**: Name, email, Interview URL, Last Interaction. **Extracted tags** (e.g. Customer-focused, Balanced approach, Resourceful). Toolbar includes **Share**, **Find Other Jobs**, **Download**, **Allow Retake** (another attempt; can override posting retake policy), and **Modular** (layout toggle) where your workspace shows them.
Across the top of the profile you'll find these tabs: **Voice Interview**, **Diagnostics**, **Verify**, **Activity**, **Notes**, and **Ask Milo**.
#### Interview URL
The **Interview URL** on the candidate card is that person's unique link — **click to copy** it. To get the public apply link shared by all candidates on an interviewer, open the interviewer → **Integrations → Inbound → Share Link**. To set or change the URL slug when you create an interviewer, see [Creating an Interviewer](creating-an-interviewer). For a custom domain on interview links, see [Custom Domains](../../outreach/outreach/custom-domains).
#### Classic vs Modular
HeyMilo has two candidate review layouts: **Classic** and **Modular**. Use **Modular** on the candidate toolbar to switch.
**What only changes in Modular**
* **Evaluations** tab: Structured modules (e.g. salary, notice period, integrity confirmations).
* **When the story doesn’t match**: Sometimes a résumé says one thing and the **voice or video interview** says another. **Cross-Agent Intelligence** (**Evaluations** tab) is HeyMilo’s way of **flagging those gaps** so you’re not the one playing detective across tabs. It’s there for **trust and accuracy**: dates, titles, skills, or claims that **don’t line up** between what they **submitted** and what they **said live**.
* If the panel is empty or errors out, use **Retry Analysis** (and **Refresh** or collapse if that’s what you see). **Nothing here auto-rejects anyone. I**t’s a **heads-up for your team** before you shortlist, pass, or follow up.
**Classic and Modular**
* **Find Other Jobs**: HeyMilo runs an **agent process** to match this person to other open roles. Use this action in the same row as **Share**, **Download**, and **Allow Retake.**
**Note:** Needs a working **ATS** connection and relevant jobs/postings for meaningful matches.
#### Candidate vs. Average
On the **Scorecard**, HeyMilo compares this candidate to other people who **completed the same interviewer**.
The diagram on the side shows live **Candidate vs Average**; hover a point for **Candidate %**, **Avg %**, and how far above or below (e.g. **32% above average**). Use it for a fast read on who’s **stronger than typical** for this role. If **N** is small, treat it as a **rough guide**.
**Right Sidebar**: **Nudge Candidate**, processed checkmarks, **Overall score**, **Communication Score**, video where shown.
#### Résumé-based questions on the call
Want the AI to **dig into their résumé** during a voice or video interview? When you build the interviewer, open the **Voice Interview** step and use **Tools.** Turn on the toggle **Resume Questions**. Keep a **résumé step before voice** in the workflow so there’s something to pull from.
So, when you **review** someone: **Voice Interview** → **Resume-Derived Questions**.
* **Upfront** = meant early,
* **Deferred** = meant later.
* **Not Asked** means the **AI didn’t get to that prompt**
### 6. Diagnostics
The Diagnostics tab shows how a candidate completed the interview — use it when a score surprises you, someone didn’t finish, or you want more context before shortlisting.
Where: Open a candidate → **Diagnostics**.
What you’ll find:
* Candidate Experience Intelligence: Overall experience score (e.g. *88 · excellent*) plus AI Analysis bullets summarizing the session (e.g. *“completed on first attempt,” “media permission errors but candidate persisted,” “slower internet detected”*).
* Visit Timeline: Each visit with status (Completed / in progress), duration, timestamp, and browser / OS (e.g. *Chrome 148 · Mac OS X*).
* Experience metrics: Scores for Journey Completion, Error Impact, Frustration Signals, Connectivity Quality, and Time to Start. Helps separate fit issues from experience issues.
* Network & session details: Connectivity chart (Avg RTT, Jitter, Packet Loss), Frustration Signals (rage click count and locations), and Candidate Rating.
* Session Replay: Play back the candidate’s screen interactions. Can take a moment to load, sensitive info is filtered, intended for internal review.
* Interview Steps: Completion status for each stage in the workflow (e.g. *Web Interview · Completed*).
* Technical details: IP Address and Evaluator Processing Times (e.g. Scorecard, English Proficiency, Highlights, Tags Extraction) for troubleshooting.
TL;DR: Scorecard answers *how did they do?* Diagnostics answers *what happened?*
### 7. Verify tab
The **Verify** tab surfaces identity checks and cheat detection signals in one place. Open it from the candidate profile alongside Diagnostics, Notes, and Activity.
Where: Open a candidate → **Verify**.
At the top of the Verify tab you will see the candidate’s location, device (OS and browser), timezone compared to yours, and VPN status.
#### Identity Verification (Beta)
Checks the candidate's email and phone to confirm their identity.
* **Email Verification**: Confirms the email address is valid and deliverable.
* **Phone Verification**: Confirms the phone number is valid and reachable.
When everything looks clean, you will see **All Checks Passed**.
#### Cheat Detection Overview (Beta)
Surfaces behavioral and environmental signals from the interview session. Flags are for human review only. Nothing here auto-rejects candidates or changes their score.
**What you will see:**
* **Risk level**: Overall assessment with a recommended action (Reject or Further Review).
* **Total Breaches**: Number of flagged events in the session.
* **% Compromised**: Share of the session where signals were detected.
* **Categories Triggered**: Which detection types were activated.
* **Severity Breakdown**: High, medium, and low signal counts with percentage of session for each.
* **Cheating Timeline**: A visual timeline of when flags occurred across the interview.
* **By Category**: A table showing category name, severity, breach count, percentage of session, and duration.
* **Breach Details**: Timestamp range, severity, confidence percentage, recommended action, the candidate's original response, and sentence-level analysis.
Flags are informational, not confirmed violations. Always review the transcript, answers, and video alongside any flags before drawing conclusions.
### Notifications
Configure alerts for completions, high scores, or expirations in Settings & Admin. For the in-app **bell** (push, sound, click-through to candidates), see [Your Dashboard](/broken/pages/ohFtbzfnPXqAQ195sfvg).
### Next steps
* [How Scoring Works](how-scoring-works) — How scores are built and how to read the report
* [Integrations](../../integrations) — ATS sync and sending results to your ATS