Back to Blog

Boolean Search for Recruiters: Operators, Strings, and Examples

November 17, 20269 min readBy: Tanner Gilligan
Boolean Search for Recruiters with the Wrangle logo on an orange and pink gradient.

Boolean search combines keywords with AND, OR, and NOT to narrow or broaden a candidate search. Recruiters use it to find profiles mentioning particular titles, skills, or certifications. Quotation marks keep phrases together; parentheses group alternatives.

It is useful when you know which words to look for. It becomes harder when the requirement is about what someone built, the problems they solved, or how their career developed. This guide covers practical strings, the differences between search platforms, and how to turn the same hiring need into a plain-language search in Wrangle.

Key takeaways

  • Use OR for alternatives and AND for separate requirements. Add exclusions only after reviewing the results.
  • Match the syntax to the platform. LinkedIn Boolean, Google X-ray, and GitHub user search are different tools.
  • A keyword match is a reason to review a profile, not proof that someone meets the requirement.
  • Test your query against a candidate you already know fits. If they are missing, check both the query and the platform's coverage.
  • For requirements that need context, write a brief and review the evidence behind the matches.

What is Boolean search in recruiting?

Boolean search lets you describe the relationship between search terms. For example:

("software engineer" OR "backend engineer") AND (Python OR Golang)

In a Boolean keyword search, this asks for at least one of the titles and at least one of the languages. It does not establish the person's current title or proficiency. The fields being searched matter: a keyword may appear in a previous job, a project, or a summary.

Recruiters use this approach in sourcing tools, resume databases, and ATS search where Boolean is supported. Support for individual operators varies, so a string that works in one tool may need changes in another.

Boolean operators: what each one does

Operator or modifierPurposeExample
ANDRequire both termsPython AND SQL
ORAccept either term, including both(Python OR Golang)
NOTExclude a termPython NOT recruiter
Quotation marksSearch for a phrase"site reliability engineer"
ParenthesesGroup alternatives and requirements("data scientist" OR "applied scientist") AND Python

Use uppercase AND, OR, and NOT on LinkedIn. Keep title alternatives inside one set of parentheses so they do not get mixed up with skill requirements.

Wildcards are platform-specific. Some databases support engineer* to match word variations. LinkedIn does not support the asterisk wildcard. Do not assume it works wherever you paste a query.

How LinkedIn, Google, and GitHub differ

PlatformWhat to useWhat to watch
LinkedIn Recruiter and Recruiter LiteBoolean keywords, quoted phrases, and structured filtersKeywords can match past experience. Use current-title or current-company filters when that distinction matters.
Google X-raysite:, quoted phrases, and - to exclude termsResults come from indexed public pages, not LinkedIn's full member database.
GitHub user searchQualifiers such as type:user, language:, and location:Repository language and self-reported location are clues, not a complete account of someone's skills or availability.

LinkedIn's filters also offer Can have, Must have, and Doesn't have choices. These correspond to OR, AND, and NOT. Avoid putting the same restriction into both a filter and a long keyword string unless you need both.

The examples below follow documented syntax. Treat them as starting points to test in your own account, with its filters and accessible data.

Boolean search strings by role

Use the first seven examples in LinkedIn Recruiter's Keywords filter. Start without broad exclusions such as NOT intern or NOT analyst: those words can appear in the work history of experienced candidates.

1. Backend software engineer

("backend engineer" OR "back-end engineer" OR "software engineer" OR "platform engineer") AND (Python OR Golang OR Rust) AND (PostgreSQL OR Kafka OR "distributed systems")

Replace the language group with the stack you need. Golang reduces the ambiguity of the word Go, but can miss profiles that use only Go. Run a separate search for that variant if it matters. A title such as "Member of Technical Staff" may need its own search too.

2. Machine learning engineer

("machine learning engineer" OR "ML engineer" OR "research engineer" OR "research scientist" OR "applied scientist") AND (PyTorch OR JAX OR TensorFlow) AND (LLM OR "large language models" OR "recommendation systems" OR "computer vision")

The last group covers several specialties. Keep the alternatives relevant to the role. Adding alternatives with OR broadens the search; making each specialty a separate AND requirement would narrow it.

3. DevOps or site reliability engineer

("site reliability engineer" OR SRE OR "platform engineer" OR "DevOps engineer" OR "infrastructure engineer") AND (Kubernetes OR Terraform) AND (AWS OR GCP OR Azure)

This finds vocabulary associated with infrastructure work. Review what the person owned: deploying an application and operating a production platform are different kinds of experience.

4. Data scientist

("data scientist" OR "applied scientist" OR "machine learning scientist") AND Python AND SQL AND ("A/B testing" OR experimentation OR "causal inference")

This example is for an experimentation-focused role. For other data science work, change that final group. Avoid NOT analyst, which can remove people who moved from analytics into data science.

5. Security engineer

("security engineer" OR "detection engineer" OR "security researcher") AND ("threat modeling" OR "incident response" OR SIEM OR "penetration testing") AND (AWS OR GCP OR Azure)

Choose the security discipline you actually need. These terms identify possible experience; they do not tell you which systems the person secured or what they personally delivered.

6. Product manager

("product manager" OR "product lead" OR "head of product") AND (B2B OR SaaS OR "enterprise software") AND (roadmap OR "product strategy" OR "go-to-market")

Use title filters if you need a current product role. Excluding "project manager" across the whole profile could remove someone who held that title earlier in their career.

7. Account executive for SaaS

("account executive" OR "enterprise account executive" OR "sales executive") AND (SaaS OR "B2B software") AND (quota OR "new business")

Avoid making "President's Club" or quota attainment a required phrase. Strong sellers may not publish those details, and a profile claim still needs checking. Discuss deal size, territory, and performance during evaluation.

8. Developers on GitHub

Use GitHub's Users search for this example:

type:user language:Rust location:Berlin

This uses GitHub qualifiers, not a LinkedIn Boolean string. It looks for personal accounts by repository language and the location in their profile. Review the actual work before deciding whether someone fits.

GitHub also supports repos: and followers: filters, but neither repository count nor follower count is a reliable hiring standard. Adding minimums can hide useful candidates.

How to build a better Boolean string

  1. Choose the essential signal. Start with the role or the experience that would make someone worth reviewing.
  2. Group alternatives with OR. Include the titles and skill names people actually use.
  3. Add requirements one at a time. Review the change in results before adding another AND group.
  4. Test exclusions carefully. Check whether a NOT term removes people you would want to contact.
  5. Check a known candidate. Confirm that their searchable profile contains your terms and is available in the platform. Then see whether the query retrieves them.

If you need a current employer, location, or title, use the platform's dedicated filters where available. A keyword anywhere in a profile does not reliably establish a current attribute.

How to use Google X-ray search

An X-ray search uses site: to focus a search engine on a particular website. For public LinkedIn profile pages, start with:

site:linkedin.com/in "backend engineer" "Kafka" "San Francisco"

Run a separate query for a title variant:

site:linkedin.com/in "platform engineer" "Kafka" "San Francisco"

You can exclude a word with a minus sign, such as -recruiter, but review the effect before keeping it. As with NOT, an exclusion can remove an otherwise relevant page.

Keep site: attached to the domain. site:linkedin.com/in works as an operator; adding a space after the colon changes the query.

Google can only return pages it has indexed. A missing profile does not mean the candidate lacks the skill, and a matching company or city may refer to past experience. Check the profile itself. There is no need to copy a complex LinkedIn string into Google and assume it will behave identically.

Common Boolean search mistakes

  • Leaving OR groups ungrouped. Operators have an order of precedence; the query is not simply read left to right. Parentheses make the intended grouping explicit.
  • Adding every requirement from the job description. One missing keyword can exclude someone worth reviewing.
  • Searching only titles. Relevant work may be described in experience, projects, or publications.
  • Using broad exclusions. NOT manager can remove someone whose summary mentions working with a manager. NOT intern can remove a senior engineer's early career history.
  • Assuming a match verifies a qualification. A certification name is a keyword. It does not establish that the certification is current.
  • Treating zero results as proof there are no candidates. Check syntax, filters, searchable fields, and coverage before drawing that conclusion.

Where Boolean search reaches its limits

Boolean is useful for explicit terms. It is less useful when the requirement depends on context.

"Built the first version of a payments system" is not a single phrase everyone uses. Neither is "moved from a large company into an early-stage team and owned infrastructure." You can approximate those requirements with keywords, but you still have to read the work history and connect the evidence.

Title variants create a similar problem. A design lead may do the work you associate with a UX designer while using different language. Expanding the OR group helps, but maintaining every possible synonym takes time.

Search from a brief in Wrangle

Wrangle Sourcing lets you start with a description of the person, a job description, or a LinkedIn profile as a calibration example.

For the backend search above, try:

Find backend engineers who use Python or Go and have owned Kafka pipelines in production at a Series A to C startup. Look for evidence that they operated those systems, including on-call responsibility.

The brief adds what matters about the experience. Wrangle evaluates candidates against the criteria and provides written reasoning to review. You can refine the criteria in the search, for example by asking for more individual-contributor experience and less management.

Wrangle also supports searching connected networks, ATS candidates, and saved candidate sources. Connect and select the relevant sources for your search; the records available depend on those connections.

Read the evidence before adding someone to your shortlist. A score prioritizes review. It does not guarantee that profile details are current or that someone is interested in a new role.

When to use a string or a brief

Use Boolean when…Use a plain-language brief when…
You need to find a specific phrase or qualification nameYou care about the work someone owned and its context
You know the vocabulary and want explicit keyword conditionsSimilar candidates use different titles or terminology
Your database supports structured keyword queriesYou want to start from a job description or an example profile

Both approaches need human review. Choose the one that helps you find and assess relevant people with less manual work.

Compare the approaches on one role

Pick a role you understand well and agree on what makes a candidate worth interviewing. Give each approach the same search time and the same hiring requirements.

As a practical starting point, spend 15 minutes on each search and review up to 25 results from each. Record fewer results if that is all a tool returns. Have a colleague review the profiles without seeing which tool found them.

Compare qualified candidates, the time it took to build the shortlist, and whether you could check why each person matched. These are suggested test sizes, not performance benchmarks. Allow additional time for the review.

For a more structured comparison, use the AI sourcing evaluation scorecard.

Frequently asked questions

What does a Boolean search look like?

("software engineer" OR "backend engineer") AND Python asks for a profile matching either title and Python. The result depends on the platform and the fields being searched.

Can you use Boolean operators in LinkedIn Recruiter?

Yes. Recruiter and Recruiter Lite support AND, OR, NOT, quoted searches, and parentheses. Use uppercase operators and check the structured filters alongside the keyword query.

What is the difference between Boolean and X-ray search?

Boolean describes the logic between terms. X-ray describes searching a particular website through a search engine, often using Google's site: operator. Use the syntax that the search engine supports.

Can you find candidates without writing Boolean strings?

Yes. In Wrangle, you can describe the person, paste a job description, or use a LinkedIn profile as a calibration example. Review the resulting criteria and candidate evidence as you refine the search.

Why does my Boolean string return zero results?

Remove a requirement, check the parentheses and operator capitalization, and review any active filters. Also check whether the candidate data you expect is searchable in that platform. A valid string can still return nothing if the data is missing.

Can an AI assistant write Boolean strings?

Yes, it can draft a starting string. Specify the platform and ask it to separate required terms from alternatives. Check the syntax and test the results before relying on it; a plausible-looking query can still exclude the people you want.

Try the plain-language version

Take one of the roles above, describe the experience that matters, and review who matches and why. Start a free Wrangle trial.

Sources and last checked date

Platform syntax checked November 17, 2026. The examples are starting queries, not claims of measured search performance.

  • LinkedIn Help, Use Boolean search on LinkedIn: https://www.linkedin.com/help/recruiter/answer/a524335
  • LinkedIn Help, Use Boolean to filter search results in Recruiter and Recruiter Lite: https://www.linkedin.com/help/linkedin/answer/a415295
  • Google Search Help, Refine Google searches.
  • GitHub Docs, Searching users.

Ready to get started?

Create an account and start running searches with your free trial, or book a demo for your team.

Level up talent operations

Find candidates and keep every search moving.

Sourcing

Start with a free trial

Get your team running in Wrangle in as little as five minutes.

Sign up