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

# Prompt pack

> Copy-ready prompts that run a full recruiting workflow: intake, market analysis, search, shortlists, outreach, and ATS rediscovery.

Copy any prompt below into a connected assistant. Every prompt uses the same example role, a Staff
Machine Learning Engineer for an AI agents team, so you can follow one req through the whole funnel.
Swap the role, locations, and companies for your own.

The prompts assume SeekOut is already connected. See [Quickstart](/mcp/getting-started/quickstart) if
it is not.

## The end-to-end workflow

Run these eight prompts in order in a single conversation. Each one builds on the search and
candidate context established by the previous prompt.

<Steps>
  <Step title="Prepare for intake and market analysis">
    Turns vague intent into a sourcing plan before you look at a single profile.

    <Prompt description={"I need to hire a Staff Machine Learning Engineer for an AI agents team. The person should have production LLM experience, strong Python, distributed systems depth, and ideally search, ranking, or recommendation systems experience.\n\nBefore we search for candidates, help me prepare for intake. Break down the role, identify must-haves vs nice-to-haves, suggest title variants, recommend target companies, and compare the talent market across Seattle, Bay Area, New York, and Toronto. Do not show candidates yet."} icon="clipboard-list">
      I need to hire a Staff Machine Learning Engineer for an AI agents team. The person should have
      production LLM experience, strong Python, distributed systems depth, and ideally search,
      ranking, or recommendation systems experience.

      Before we search for candidates, help me prepare for intake. Break down the role, identify
      must-haves vs nice-to-haves, suggest title variants, recommend target companies, and compare the
      talent market across Seattle, Bay Area, New York, and Toronto. Do not show candidates yet.
    </Prompt>
  </Step>

  <Step title="Run the search">
    Returns people with the match signals, not just names.

    <Prompt description="Now run the SeekOut search for this role. Focus on Staff-level ML Engineers, AI Infrastructure Engineers, Applied ML Engineers, and similar titles. Prioritize candidates in Seattle or the Bay Area, but include remote-friendly candidates if they are very strong. Look for production LLM systems, Python, distributed systems, search, ranking, recommendations, or agent infrastructure experience. Show me the top candidates and explain the match signals." icon="search">
      Now run the SeekOut search for this role. Focus on Staff-level ML Engineers, AI Infrastructure
      Engineers, Applied ML Engineers, and similar titles. Prioritize candidates in Seattle or the Bay
      Area, but include remote-friendly candidates if they are very strong. Look for production LLM
      systems, Python, distributed systems, search, ranking, recommendations, or agent infrastructure
      experience. Show me the top candidates and explain the match signals.
    </Prompt>
  </Step>

  <Step title="Evaluate and rank the strongest fits">
    Produces a hiring-manager-ready shortlist with evidence and gaps.

    <Prompt description="Evaluate the top candidates against the role. Rank them by fit and give me clear reasoning for each one. For every candidate, explain: why they are a strong fit; what evidence supports the match; any concerns or gaps; and what I should ask them in a first conversation. Give me a hiring-manager-ready shortlist." icon="scale">
      Evaluate the top candidates against the role. Rank them by fit and give me clear reasoning for
      each one. For every candidate, explain: why they are a strong fit; what evidence supports the
      match; any concerns or gaps; and what I should ask them in a first conversation. Give me a
      hiring-manager-ready shortlist.
    </Prompt>
  </Step>

  <Step title="Add the top 25 to a workspace">
    Moves the shortlist out of the conversation and into SeekOut.

    <Prompt description="Add the top 25 candidates to a new SeekOut workspace called “Staff ML Engineer - AI Agents.” Use the current search and candidate ranking. After adding them, give me the workspace link so I can review the shortlist and search directly in SeekOut." icon="folder-plus">
      Add the top 25 candidates to a new SeekOut workspace called “Staff ML Engineer - AI Agents.” Use
      the current search and candidate ranking. After adding them, give me the workspace link so I can
      review the shortlist and search directly in SeekOut.
    </Prompt>
  </Step>

  <Step title="Fetch contact details">
    Retrieves contacts with a confirmation step before credits are used.

    <Prompt description="Fetch available contact details for the top 25 candidates in the “Staff ML Engineer - AI Agents” workspace. Before using credits, summarize how many candidates you will retrieve contact details for and ask me to confirm." icon="contact">
      Fetch available contact details for the top 25 candidates in the “Staff ML Engineer - AI Agents”
      workspace. Before using credits, summarize how many candidates you will retrieve contact details
      for and ask me to confirm.
    </Prompt>
  </Step>

  <Step title="Draft personalized outreach for the top 10">
    Outreach grounded in each candidate's real background.

    <Prompt description="Help me draft personalized outreach for the top 10 candidates. For each candidate, write a short email that references something specific from their background, explains why this AI agents role may be interesting, and sounds like a real recruiter wrote it. Avoid generic AI language. Keep each message concise and easy to personalize." icon="send">
      Help me draft personalized outreach for the top 10 candidates. For each candidate, write a short
      email that references something specific from their background, explains why this AI agents role
      may be interesting, and sounds like a real recruiter wrote it. Avoid generic AI language. Keep
      each message concise and easy to personalize.
    </Prompt>
  </Step>

  <Step title="Expand the pool at the same bar">
    Widens the funnel without lowering the technical requirements.

    <Prompt description="Show me how to expand this candidate pool without lowering the bar. Suggest alternate titles, adjacent skills, feeder companies, nearby locations, remote markets, and different candidate backgrounds we may be missing. Then recommend the best 3 expansion strategies and explain the tradeoffs." icon="git-branch">
      Show me how to expand this candidate pool without lowering the bar. Suggest alternate titles,
      adjacent skills, feeder companies, nearby locations, remote markets, and different candidate
      backgrounds we may be missing. Then recommend the best 3 expansion strategies and explain the
      tradeoffs.
    </Prompt>
  </Step>

  <Step title="Rediscover candidates in your ATS">
    Surfaces talent you already paid to attract.

    <Prompt description="Now search our ATS for candidates who could fit this Staff ML Engineer role. Look for past applicants, silver medalists, or candidates already in our ATS with ML infrastructure, LLM, Python, distributed systems, search, ranking, or recommendation systems experience. Prioritize candidates who reached later stages, were rejected for timing or non-skill reasons, or now appear stronger based on updated experience. Give me a ranked list of rediscovery candidates and explain why each one is worth revisiting." icon="archive">
      Now search our ATS for candidates who could fit this Staff ML Engineer role. Look for past
      applicants, silver medalists, or candidates already in our ATS with ML infrastructure, LLM,
      Python, distributed systems, search, ranking, or recommendation systems experience. Prioritize
      candidates who reached later stages, were rejected for timing or non-skill reasons, or now
      appear stronger based on updated experience. Give me a ranked list of rediscovery candidates and
      explain why each one is worth revisiting.
    </Prompt>
  </Step>
</Steps>

## Recruiter productivity

<h3 id="vague-intent-to-sourcing-plan">
  Turn vague hiring intent into a sourcing plan
</h3>

<Prompt description="I need to hire a Staff Machine Learning Engineer for our AI agents team. They need deep Python, LLM application experience, production ML systems, and ideally experience with search, ranking, or recommendation systems. We prefer Seattle or Bay Area, but we can consider remote. Build the sourcing strategy, identify title variants, must-have and nice-to-have criteria, target companies, run the search, verify the candidate fit, and give me the first 10 strongest candidates with reasons." icon="target">
  I need to hire a Staff Machine Learning Engineer for our AI agents team. They need deep Python, LLM
  application experience, production ML systems, and ideally experience with search, ranking, or
  recommendation systems. We prefer Seattle or Bay Area, but we can consider remote. Build the
  sourcing strategy, identify title variants, must-have and nice-to-have criteria, target companies,
  run the search, verify the candidate fit, and give me the first 10 strongest candidates with
  reasons.
</Prompt>

<h3 id="hidden-ai-talent">
  Find hidden AI talent across public, GitHub, and academic signals
</h3>

<Prompt description="Find rising-star AI infrastructure engineers who have evidence across GitHub, academic work, or public profiles. Focus on people with LLM systems, retrieval, evals, distributed inference, or agent infrastructure experience. Search across all relevant SeekOut verticals, dedupe the results, and rank candidates by evidence strength, not just title." icon="sparkles">
  Find rising-star AI infrastructure engineers who have evidence across GitHub, academic work, or
  public profiles. Focus on people with LLM systems, retrieval, evals, distributed inference, or agent
  infrastructure experience. Search across all relevant SeekOut verticals, dedupe the results, and
  rank candidates by evidence strength, not just title.
</Prompt>

<h3 id="find-similar-candidates">
  Find more people like this candidate
</h3>

<Prompt description="I like this candidate profile because they have startup experience, strong backend systems depth, and recent LLM product work. Reverse-engineer what makes them a strong fit, then find 15 similar candidates. Avoid clones from the same company unless they are exceptional." icon="user-search">
  I like this candidate profile because they have startup experience, strong backend systems depth,
  and recent LLM product work. Reverse-engineer what makes them a strong fit, then find 15 similar
  candidates. Avoid clones from the same company unless they are exceptional.
</Prompt>

<h3 id="calibrate-from-feedback">
  Calibrate from hiring-manager feedback
</h3>

<Prompt description="Here is my feedback on the slate: candidate 1 is too enterprise SaaS, candidate 2 is great because of hands-on infra depth, candidate 3 is too research-heavy, and candidate 4 is close but too junior. Refine the search and bring me a better slate of 10 candidates." icon="sliders-horizontal">
  Here is my feedback on the slate: candidate 1 is too enterprise SaaS, candidate 2 is great because
  of hands-on infra depth, candidate 3 is too research-heavy, and candidate 4 is close but too junior.
  Refine the search and bring me a better slate of 10 candidates.
</Prompt>

<h3 id="hiring-manager-shortlist">
  Build a hiring-manager-ready shortlist
</h3>

<Prompt description="Build a ranked shortlist of the top 5 candidates for this role. For each person, include why they fit, what evidence supports the match, concerns or gaps, suggested interview focus areas, and a one-line hiring manager summary." icon="list-ordered">
  Build a ranked shortlist of the top 5 candidates for this role. For each person, include why they
  fit, what evidence supports the match, concerns or gaps, suggested interview focus areas, and a
  one-line hiring manager summary.
</Prompt>

<h3 id="outreach-after-discovery">
  Draft personalized outreach after discovery
</h3>

<Prompt description="For the top 5 candidates in this slate, draft personalized outreach. Make each message specific to their background, avoid generic AI-sounding language, and give me one short LinkedIn version and one email version per person." icon="message-square-text">
  For the top 5 candidates in this slate, draft personalized outreach. Make each message specific to
  their background, avoid generic AI-sounding language, and give me one short LinkedIn version and one
  email version per person.
</Prompt>

## Talent market and leadership intelligence

<h3 id="competitive-talent-map">
  Build a competitive talent map
</h3>

<Prompt description="Map the AI platform and applied ML talent at Anthropic, OpenAI, Google DeepMind, Meta, and Databricks. I want to understand titles, seniority, skills, geographic distribution, likely feeder companies, and which candidate segments are most reachable for a startup." icon="map">
  Map the AI platform and applied ML talent at Anthropic, OpenAI, Google DeepMind, Meta, and
  Databricks. I want to understand titles, seniority, skills, geographic distribution, likely feeder
  companies, and which candidate segments are most reachable for a startup.
</Prompt>

<h3 id="compare-markets">
  Compare markets before opening a req
</h3>

<Prompt description="Compare Seattle, San Francisco, New York, Toronto, and London for Staff-level ML infrastructure engineers with Python, distributed systems, and LLM production experience. Show talent pool size, top employers, common titles, skill concentration, and which market gives us the best odds." icon="columns-3">
  Compare Seattle, San Francisco, New York, Toronto, and London for Staff-level ML infrastructure
  engineers with Python, distributed systems, and LLM production experience. Show talent pool size,
  top employers, common titles, skill concentration, and which market gives us the best odds.
</Prompt>

<h3 id="academic-expert-discovery">
  Discover academic and industry experts
</h3>

<Prompt description="Find academic and industry experts in AI evaluation, LLM benchmarking, and agent reliability. Prioritize people with publications, conference signals, or strong research depth. Give me a ranked expert map and identify who might be reachable for an industry role." icon="graduation-cap">
  Find academic and industry experts in AI evaluation, LLM benchmarking, and agent reliability.
  Prioritize people with publications, conference signals, or strong research depth. Give me a ranked
  expert map and identify who might be reachable for an industry role.
</Prompt>

<h3 id="bench-depth-analysis">
  Analyze bench depth
</h3>

<Prompt description="Analyze our bench depth for three critical roles: Principal ML Engineer, Staff Backend Engineer, and Product Manager for AI Recruiting. Show who internally has overlapping skills, where we have gaps, and which roles are most exposed if someone leaves." icon="layers">
  Analyze our bench depth for three critical roles: Principal ML Engineer, Staff Backend Engineer, and
  Product Manager for AI Recruiting. Show who internally has overlapping skills, where we have gaps,
  and which roles are most exposed if someone leaves.
</Prompt>

<h3 id="pipeline-health-snapshot">
  Take a pipeline health snapshot
</h3>

<Prompt description="Give me a pipeline overview for our active engineering reqs. Break down candidates by stage, recruiter, rejection reason, and timeline. Highlight bottlenecks, stale stages, and where we should intervene this week." icon="activity">
  Give me a pipeline overview for our active engineering reqs. Break down candidates by stage,
  recruiter, rejection reason, and timeline. Highlight bottlenecks, stale stages, and where we should
  intervene this week.
</Prompt>

<h3 id="inclusive-sourcing-expansion">
  Expand sourcing inclusively
</h3>

<Prompt description="Analyze this talent pool for diversity and representation signals. Then suggest ways to broaden the search without lowering the bar: alternate titles, adjacent companies, schools, geographies, and non-obvious candidate backgrounds." icon="users-round">
  Analyze this talent pool for diversity and representation signals. Then suggest ways to broaden the
  search without lowering the bar: alternate titles, adjacent companies, schools, geographies, and
  non-obvious candidate backgrounds.
</Prompt>

<h3 id="comp-market-reality-check">
  Run a compensation and market reality check
</h3>

<Prompt description="For this Senior Product Manager, AI Recruiting role in Bellevue, give me compensation context and market difficulty. Compare Bellevue, San Francisco, New York, and remote. Tell me whether our likely range is competitive and where we may struggle." icon="dollar-sign">
  For this Senior Product Manager, AI Recruiting role in Bellevue, give me compensation context and
  market difficulty. Compare Bellevue, San Francisco, New York, and remote. Tell me whether our likely
  range is competitive and where we may struggle.
</Prompt>

## Power filters and vertical search

<h3 id="federal-cleared-sales-talent">
  Source federal-cleared sales talent
</h3>

<Prompt description="Find Federal Account Executives in the DC metro area with security clearance signals, experience selling into federal civilian or defense agencies, and prior work at defense contractors or federal SaaS vendors. Show me the best candidates and explain which clearance or federal-sales signals you found." icon="shield-check">
  Find Federal Account Executives in the DC metro area with security clearance signals, experience
  selling into federal civilian or defense agencies, and prior work at defense contractors or federal
  SaaS vendors. Show me the best candidates and explain which clearance or federal-sales signals you
  found.
</Prompt>

<h3 id="healthcare-specialist-search">
  Search for healthcare specialists
</h3>

<Prompt description="Find cardiologists licensed in California who have experience with heart failure or electrophysiology and are affiliated with major hospital systems. Prioritize people in the Bay Area or Los Angeles, but show me how much the pool expands statewide." icon="stethoscope">
  Find cardiologists licensed in California who have experience with heart failure or
  electrophysiology and are affiliated with major hospital systems. Prioritize people in the Bay Area
  or Los Angeles, but show me how much the pool expands statewide.
</Prompt>

<h3 id="nursing-licensing-search">
  Search nursing talent with licensing constraints
</h3>

<Prompt description="Find nurse practitioners in Texas with emergency medicine or urgent care experience. Include license-state signals, specialties, likely current employers, and broaden the search carefully if the pool is too small." icon="heart-pulse">
  Find nurse practitioners in Texas with emergency medicine or urgent care experience. Include
  license-state signals, specialties, likely current employers, and broaden the search carefully if
  the pool is too small.
</Prompt>

## Agentic execution

<h3 id="internal-talent-redeployment">
  Redeploy internal talent
</h3>

<Prompt description="Search our internal talent for people who could move into an AI solutions engineering role. Look for employees with customer-facing experience, Python or ML exposure, strong product judgment, and recruiting-domain knowledge. Group them by readiness: ready now, ready with training, and long-shot." icon="arrow-right-left">
  Search our internal talent for people who could move into an AI solutions engineering role. Look for
  employees with customer-facing experience, Python or ML exposure, strong product judgment, and
  recruiting-domain knowledge. Group them by readiness: ready now, ready with training, and long-shot.
</Prompt>

<h3 id="rediscover-silver-medalists">
  Rediscover silver medalists in the ATS
</h3>

<Prompt description="Search our ATS for past candidates who could fit a Senior Backend Engineer role today. Focus on people who reached onsite or final stages, were rejected for timing or compensation reasons, and now have more relevant experience. Rank the best rediscovery candidates." icon="refresh-cw">
  Search our ATS for past candidates who could fit a Senior Backend Engineer role today. Focus on
  people who reached onsite or final stages, were rejected for timing or compensation reasons, and now
  have more relevant experience. Rank the best rediscovery candidates.
</Prompt>

<h3 id="contact-and-export-workflow">
  Run a contact and export workflow
</h3>

<Prompt description="For the top 20 candidates, retrieve available emails, add them to a workspace called “Staff ML Infra Q3,” and prepare them for export to our ATS. Before using credits or exporting, summarize what will happen and ask for confirmation." icon="upload">
  For the top 20 candidates, retrieve available emails, add them to a workspace called “Staff ML Infra
  Q3,” and prepare them for export to our ATS. Before using credits or exporting, summarize what will
  happen and ask for confirmation.
</Prompt>

<h3 id="end-to-end-single-request">
  Run the full workflow in one request
</h3>

<Prompt description="We need to hire a Staff Machine Learning Engineer for an AI recruiting agents team. The person needs production LLM experience, Python, distributed systems, search/ranking/recommendation experience, and strong product instincts. Seattle or Bay Area preferred, remote acceptable. Start by turning this into a sourcing strategy. Then compare Seattle, Bay Area, New York, and Toronto. Search public, GitHub, and academic sources. Build a ranked shortlist of 10 candidates with evidence. Identify which 5 I should contact first, draft personalized outreach, create a workspace, and prepare the candidates for ATS export after I approve." icon="workflow">
  We need to hire a Staff Machine Learning Engineer for an AI recruiting agents team. The person needs
  production LLM experience, Python, distributed systems, search/ranking/recommendation experience,
  and strong product instincts. Seattle or Bay Area preferred, remote acceptable. Start by turning
  this into a sourcing strategy. Then compare Seattle, Bay Area, New York, and Toronto. Search public,
  GitHub, and academic sources. Build a ranked shortlist of 10 candidates with evidence. Identify
  which 5 I should contact first, draft personalized outreach, create a workspace, and prepare the
  candidates for ATS export after I approve.
</Prompt>

## Related

* [Prompting tips](/mcp/capabilities/prompting-patterns) — how to structure and refine your own requests
* [Recruiting workflows](/mcp/capabilities/workflows) — what SeekOut MCP can do, by task
* [Work with candidates](/mcp/capabilities/candidate-actions) — credit usage and write effects for the action prompts above