How can we help?

RoleMule — one mule for every role. It turns a raw job posting into a complete application package in about 30 seconds. For every job you analyze, you get ten tabs of AI-generated insights:

• Company — Culture, funding stage, interview culture, and insider tips
• Your Fit — A match score with a breakdown of strengths and gaps
• Strategy — How to position yourself and what to emphasize in your application
• Job Details — Extracted salary, work type, required skills, qualifications, and all listed office locations
• Cover Letter — A tailored letter referencing your specific experience
• Resume — Personalized recommendations to optimize your resume for this role
• Interview Prep — Predicted questions, STAR answer frameworks, and process tips
• Optimize CV — Iterative AI loop that scores and rewrites your CV for the role, then produces an optimized CV and cover letter
• Mock Session — Timed conversational mock session (HR / Pro / Manager) with voice or typed answers
• Outreach — Public web contact suggestions and copy-ready outreach drafts (you send messages yourself)

Beyond job analysis, RoleMule also includes six Career Tools for follow-up emails, salary negotiation, thank you notes, job comparison, reference requests, and rejection analysis. Power users can drive the same workflows from the optional CLI (rolemule) — including from Claude or Cursor.

1. Create an account — Sign up with email/password or continue with Google
2. Complete your profile — Upload your resume to auto-fill your experience, education, and skills, or fill it in manually. Either way, review and adjust to make sure everything is accurate. The more complete your profile, the more personalized the AI output will be
3. Choose an AI provider and add an API key — In Settings → AI Setup, pick Gemini, OpenAI, Anthropic, or local Ollama (no key). See the API Key Setup section below
4. Analyze your first job — Click New Application on the dashboard, paste a job description, and hit Analyze

Tip: A complete profile is essential — the AI uses your experience, education, and skills to personalize every output. Make sure everything looks right before your first analysis.

Yes — the RoleMule platform itself is completely free and open source. To power the AI features, you bring your own API key for Gemini, OpenAI, or Anthropic (or use local Ollama with no key) and pay the provider directly for usage. In practice, a single job analysis costs well under $0.01 on cloud providers, so personal use costs are negligible.

The AI output is only as good as the inputs. Here's how to get the most out of RoleMule:

• Make sure your profile is complete — Whether you uploaded a resume or filled it in manually, review that your experience includes specific job titles, technologies, metrics, and accomplishments — not just generic descriptions
• Give the AI the full job description — Paste it directly, upload a file, or use the Chrome extension to extract it from any job site. Include everything: responsibilities, requirements, benefits, and company blurb. More context = better analysis
• Review and edit AI output — Treat everything as a strong first draft. The cover letter and resume tips are personalized, but you know your story best
• Use the match score as a filter — A score below your threshold is a signal to reconsider or address gaps before applying. You can configure your minimum threshold in Settings → Preferences
• Read the Strategy tab — This often contains the most actionable insight, specific to how your background intersects with this particular role

1. Open Settings → AI Setup
2. Select a provider: Google Gemini, OpenAI, Anthropic, or Ollama (local)
3. For cloud providers, use Get API Key to open the provider’s console, create a key, paste it, and click Save
4. Optionally pick a preferred model for that provider

Key pages: Google AI Studio · OpenAI · Anthropic. For Ollama, pull a model on the server (e.g. ollama pull qwen3) — no API key needed.

Important: Your API key is like a password — keep it private and never share it. If it may be compromised, revoke it in the provider console and save a new one in RoleMule.

A full job analysis typically costs well under $0.01 on cloud providers. You would need to analyze hundreds of jobs per day to incur meaningful charges. Local Ollama has no per-token API bill.

Yes. Your API key is protected at every step:

• Encrypted at rest using Fernet symmetric encryption before being written to the database
• Never returned in API responses — you can set a new key but cannot read back the stored one
• Never logged in application logs or error traces
• Used only to call your chosen AI provider on your behalf — we make no other use of it
• Deletable at any time from Settings → AI Setup

When you submit a job posting, a pipeline of five specialized AI agents runs — each building on the previous one's output:

1. Job Analyzer — Extracts salary, location, work type, required skills, and qualifications from the raw posting
2. Profile Matcher — Compares your profile against job requirements, produces a match score with a detailed breakdown, and recommends how to position yourself for this specific role
3. Company Researcher — Researches the company's culture, growth stage, values, and interview style
4. Cover Letter Writer — Generates a tailored cover letter using your actual experience and the company's priorities
5. Resume Advisor — Identifies gaps and suggests targeted, high-impact changes to your resume

The full pipeline takes about 30 seconds. Interview Prep, CV Optimization, Mock Session, and Outreach can be run separately on demand from within any completed application.

Career Tools are standalone AI tools for the stages of job searching that happen outside of the initial application. Access them from Dashboard → Career Tools.

• Thank You Note — Post-interview thank you emails that reference specific conversation points
• Follow-up Email — Professional follow-ups for any stage: after applying, after an interview, or when you've been ghosted
• Salary Coach — Negotiation scripts, pushback responses, and alternative asks to maximize your offer
• Job Comparison — Compare 2–3 offers side-by-side with AI scoring on compensation, growth potential, culture fit, and more
• Reference Request — Professional emails asking former colleagues or managers to serve as references
• Rejection Analysis — Turn rejection emails into actionable feedback and improvement suggestions

CV Optimization is available on the Optimize CV tab after a job analysis workflow completes. It runs an iterative loop:

1. An AI Hiring Manager evaluates your CV against the job description and returns a score (out of 10) plus feedback
2. An AI Applicant revises the CV based on that feedback
3. The cycle repeats until the score reaches your threshold, hits the max iteration count, or stops improving
4. A final cover letter is generated to match the best-scoring CV

You can configure max iterations (2–7) and the score threshold (7.0–9.5) before starting. When you click Start Optimization, the setup panel hides and progress appears. Results include the optimized CV, cover letter, gap analysis (skills that could not be fixed by rephrasing alone), and a full iteration history. Copy buttons and Download CV are available when complete.

Download CV runs on the RoleMule server — you do not need LibreOffice on your computer. Docker deployments return a formatted .odt file (Gemini HTML → layout normalizer → LibreOffice). Local dev without LibreOffice returns a Word .docx with proper bullets and section layout. Rate limits and quota errors show as toasts, not as a markdown file download.

If the loop stops early due to API quota, the best CV reached is saved; a banner explains what happened and the cover letter section shows an empty state if it could not be generated.

Requires an AI provider configured in Settings → AI Setup (or Vertex AI if your operator enabled it). Rate limit: 10 optimization runs per hour, 10 CV downloads per hour.

Interview Prep is generated on demand for any completed job analysis. Open an application and click the Interview tab, then hit Generate. It produces three sub-sections:

• Process — What the interview process typically looks like at this company, what to expect at each stage, and how to prepare
• Questions — Predicted interview questions (behavioral and role-specific), with suggested answer frameworks using your own experience
• Preparation — A tailored checklist of what to review, practice, and prepare before your interview

You can regenerate Interview Prep if you want a fresh set of questions or if your preparation evolves.

This is different from Mock Session (the next tab), which is a live timed conversation — not a static Q&A guide.

Mock Session is a timed conversational practice on the Mock Session tab (after Interview). It uses your own API key from Settings → AI Setup.

1. Choose a style — HR, Pro, or Manager — and a duration of 10, 15, or 20 minutes
2. Answer by voice (browser speech recognition) or typing; the interviewer speaks back with browser text-to-speech
3. While the AI is composing its next line, you may see Interviewer is typing… as the question streams in
4. When time runs out (or you end early), you get a scored debrief with strengths and improvements

English only for now. No audio is uploaded to the server — speech stays in your browser. Requires an AI provider in Settings → AI Setup. Rate limits: 10 starts / hour, 40 turns / hour, 10 finishes / hour.

The match score is a useful signal, not a definitive verdict. It's calculated by comparing your profile against the job's stated requirements across:

• Skills and technologies (required vs. preferred)
• Years of experience
• Education requirements
• Location and work arrangement preferences
• Seniority level alignment

A high score (>75%) means you meet most stated requirements — worth applying confidently. A moderate score (50–75%) means gaps exist but the role may still be reachable — read the Strategy tab for how to address them. A low score (<50%) is a signal to reconsider whether the role is right for your current profile.

You can configure the minimum threshold at which the workflow pauses and asks you to confirm in Settings → Preferences. The default is 50%.

The score improves with a more complete profile — sparse profiles produce less reliable matching.

The four cards at the top of your dashboard summarize your pipeline:

• Total Applications — every job you've added and analyzed
• Applied — applications you've moved into the funnel (counts Applied, Interview, Offer, and Rejected tracking stages)
• Interviews — applications where you've marked Interview
• Response Rate — among tracked applications only: (Interview + Offer + Rejected) ÷ (Applied + Interview + Offer + Rejected). Cards still in analysis-only state (no tracking button selected) are excluded.

Use the Applied / Interview / Offer / Rejected buttons on each card to keep these stats accurate.

When you submit a job from the Chrome extension (or another path that saves a posting URL), the application detail page header shows a View posting link that opens the original listing in a new tab. The link appears only for valid http(s) URLs.

If the posting lists multiple office locations, the header location line shows them separated by | (primary city first, then additional offices). The Job Details tab lists the same locations.

Yes. Both the Cover Letter and Resume tabs have a Regenerate button that will produce a fresh version using the same inputs. This is useful if the initial output missed the mark. Before regenerating, you can adjust the output style in Settings → Preferences: cover letter tone (professional, conversational, or enthusiastic) and resume advice style (concise or detailed). Regeneration is rate-limited to 5 times per hour per application.

1. Locate the /extension folder inside the repository
2. Open Chrome and go to chrome://extensions
3. Enable Developer Mode (toggle in the top right)
4. Click "Load unpacked" and select the /extension folder
5. The RoleMule icon will appear in your Chrome toolbar

Note: Make sure you're logged into RoleMule in your browser — the extension uses your account for analysis and for form match.

The extension auto-detects job postings on:

• Any job board — individual job listing pages
• Company careers pages — if the job description is visible on the page, the extension can usually extract it

If a site isn't supported, use the New Application page on the dashboard — you can paste the job description directly or upload a file.

1. Browse to any job posting on a supported site
2. Click the RoleMule icon in your Chrome toolbar
3. Click "Analyze This Job" to send it straight to your dashboard
4. The AI analysis starts automatically — results in 30 seconds

1. Complete Profile Setup first — especially work authorization, visa sponsorship, and an uploaded resume if the form asks for a file
2. Browse to an application form with visible fields (main page only; fields inside embedded frames are not scanned)
3. Click the RoleMule icon in your Chrome toolbar
4. Click "Match Form To Profile" — the extension scans the form, maps fields to your profile on the server (AI plus deterministic rules for common screening questions), and applies values on the page
5. Review every field before you submit; the popup reports how many fields were filled or failed

Screening answers for work authorization and visa sponsorship come from your profile settings, not from the AI model alone.

This means you haven’t chosen a provider and added credentials yet — required before you can run any analysis (unless the operator enabled Vertex AI).

1. Go to Settings → AI Setup
2. Select a provider, add your API key (or pick Ollama), and click Save

If you don’t have a key yet, see the API Key Setup section above.

A full analysis normally completes in around 30 seconds. If it's taking longer:

• Wait up to 60 seconds — very detailed job postings can take longer
• Check your connection — the app reconnects automatically if the connection drops, but if the UI appears frozen a full page refresh will reset it
• Check your API key — an invalid or expired key will cause the analysis to fail. Go to Settings and re-save your key

RoleMule enforces per-user rate limits to prevent accidental runaway usage of your AI provider quota. The error message will tell you exactly how long to wait before trying again.

• Job analyses: 10 per hour
• Career Tools: 10 per hour
• Cover Letter / Resume regeneration: 5 per hour per application
• Interview Prep generation: 5 per hour
• CV Optimization: 10 runs per hour, 10 CV downloads per hour
• Mock Session: 10 starts / hour, 40 turns / hour, 10 finishes / hour
• Hiring Outreach: 5 generations / hour

Limits reset on a rolling hourly basis. If you are the instance operator and want to adjust these limits, they are configurable in the backend rate-limit settings.

This almost always means your profile needs more detail. The AI can only personalize output based on what's in your profile — if it's sparse, the output will be generic.

• Write a strong Professional Summary — this is the single most impactful field; describe your background, key strengths, and what you're looking for in 3–4 sentences
• Add specific job titles and companies in your work history, not just "Software Engineer"
• Include concrete accomplishments with numbers: "Reduced API latency by 40%", "Led a team of 5 engineers"
• List your actual skills, especially technologies and tools relevant to roles you're targeting
• Complete your Career Preferences — salary range, job type, and work arrangement all help tailor the output

After improving your profile, re-run the analysis on the same job posting to see the difference.

There are two ways to get job postings into RoleMule:

• Paste or upload — In the New Application form, either paste the full job description or upload a PDF, TXT, or Word (.docx) file. Include everything: title, company, responsibilities, and requirements
• Chrome Extension — On a posting page, click the RoleMule icon and choose Analyze This Job to send the description for analysis. See Chrome Extension setup for installation steps

The more complete the job text, the better the analysis. Short or incomplete descriptions produce less accurate results.

The rolemule CLI is installed with the project (make setup / pip install -e ".[cli]"). It covers the full product surface from the terminal — not just job analysis.

Quick start (one common path among many):

1. rolemule auth login — sign in (or use a personal access token for automation)
2. rolemule workflow analyze job.txt --wait — run the full analysis pipeline
3. rolemule workflow results SESSION --section cover-letter — pull outputs

Other command groups include apps (list/manage applications), interview, cv (Optimize CV), tools (career tools), profile, extension, and config. Claude, Cursor, and similar agents can call the same commands.

Run rolemule --help (or rolemule <group> --help) for the full command list, or see the repo CLI docs.

1. Go to Dashboard → Settings
2. Scroll to the Danger Zone section at the bottom
3. Click Delete Account and confirm

This is permanent and cannot be undone. All your data — profile, applications, generated content, and your API key — will be permanently deleted.

1. Go to Dashboard → Settings → Data Export
2. Click Export My Data
3. A JSON file will be downloaded to your device

The export includes your full profile, all job applications and their AI-generated content, and your workflow history. It does not include your encrypted API key.

This application stores only what's necessary to function:

• Account information — Email address and name (from registration or Google OAuth)
• Profile data — Work experience, education, skills, and career preferences that you enter
• Job application data — The job postings you analyze and the AI-generated content produced for them
• API key — Stored encrypted; used only to call your chosen AI provider on your behalf

Your data is not sold, used for advertising, or shared with third parties beyond what's required to operate the application. Read the full Privacy Policy for details.

Password: Go to Dashboard → Settings → Account and use the Change Password form. You'll need to enter your current password to confirm.

Email: Email address changes are not currently supported through self-service. If you need to update your email, you can delete your account and re-register, or open an issue on the project's GitHub repository.

If you signed in with Google OAuth, your email is managed by your Google account and cannot be changed within RoleMule.