--- title: "Keep hiring standards and candidate notes in files so AI works from evidence" description: "Store hiring standards, open roles, and structured candidate notes in files, so AI drafts postings, questions, and comparisons from your record." canonical: "https://scalewithsearch.com/articles/ai-memory-for-recruiting" date: "2026-01-28" modified: "2026-09-25" --- ## Site navigation - [Scale With Search](https://scalewithsearch.com/) - Real estate - Real estate - [Real estate](https://scalewithsearch.com/for/real-estate) - Work - Start here - [Send your brief](https://scalewithsearch.com/work#send-your-brief) - [Prepare your six-question brief](https://scalewithsearch.com/work#prepare-your-six-question-brief) - Build - [Site build, content library with SEO, signal desk](https://scalewithsearch.com/work) - For your business - Trades and home services - [Auto body and collision shops](https://scalewithsearch.com/for/auto-body-and-collision-shops) - [Foundation and home repair contractors](https://scalewithsearch.com/for/foundation-and-home-repair) - [Garage door and fencing contractors](https://scalewithsearch.com/for/garage-door-and-fencing-contractors) - [HVAC contractors](https://scalewithsearch.com/for/hvac-contractors) - [Janitorial and commercial cleaning companies](https://scalewithsearch.com/for/janitorial-and-commercial-cleaning) - [Locksmiths](https://scalewithsearch.com/for/locksmiths) - [Moving companies](https://scalewithsearch.com/for/moving-companies) - [Pest control companies](https://scalewithsearch.com/for/pest-control-companies) - [Plumbing and electrical contractors](https://scalewithsearch.com/for/plumbing-and-electrical-contractors) - [Restoration and water or fire damage companies](https://scalewithsearch.com/for/restoration-and-water-fire-damage) - [Roofing companies](https://scalewithsearch.com/for/roofing-companies) - [Towing companies](https://scalewithsearch.com/for/towing-companies) - [Tree services and landscaping companies](https://scalewithsearch.com/for/tree-services-and-landscaping) - [Solar installers](https://scalewithsearch.com/for/solar-installation) - [General contractors](https://scalewithsearch.com/for/general-contractors-and-construction) - [Paving, concrete, and flooring contractors](https://scalewithsearch.com/for/paving) - Practices and professional services - [Bookkeeping and tax practices](https://scalewithsearch.com/for/bookkeeping-and-tax-practices) - [Dental practices](https://scalewithsearch.com/for/dental-practices) - [Family and criminal defense law firms](https://scalewithsearch.com/for/family-and-criminal-defense-law-firms) - [Med spas and aesthetics practices](https://scalewithsearch.com/for/med-spas-and-aesthetics) - [Personal injury law firms](https://scalewithsearch.com/for/personal-injury-law-firms) - [Veterinary clinics](https://scalewithsearch.com/for/veterinary-clinics) - [Gyms and fitness studios](https://scalewithsearch.com/for/fitness) - [Therapy and outpatient health practices](https://scalewithsearch.com/for/therapy-and-outpatient-health) - [Medical billing companies](https://scalewithsearch.com/for/medical-billing) - [Insurance agencies](https://scalewithsearch.com/for/insurance-agencies) - [Financial advisors](https://scalewithsearch.com/for/financial-advisors) - [Property management companies](https://scalewithsearch.com/for/property-management) - [Recruiting and staffing agencies](https://scalewithsearch.com/for/recruiting-and-staffing) - [Architects and interior designers](https://scalewithsearch.com/for/architects-and-interior-designers) - [Logistics and supply chain companies](https://scalewithsearch.com/for/logistics-and-supply-chain) - Agencies, MSPs, and manufacturing - [IT and managed service providers](https://scalewithsearch.com/for/it-and-managed-service-providers) - [Machine shops and precision manufacturers](https://scalewithsearch.com/for/machine-shops-and-precision-manufacturing) - [Marketing agencies and freelancers](https://scalewithsearch.com/for/marketing-agencies-and-freelancers) - [SEO agencies and consultants](https://scalewithsearch.com/for/seo-agencies-and-consultants) - [Small manufacturers and fabricators](https://scalewithsearch.com/for/small-manufacturers-and-fabricators) - Restaurants, shops, studios, and nonprofits - [Restaurants and hospitality businesses](https://scalewithsearch.com/for/restaurants-and-hospitality) - [Retail stores and ecommerce sellers](https://scalewithsearch.com/for/retail-and-ecommerce) - [Photographers, event planners, and travel agents](https://scalewithsearch.com/for/photographers) - [Churches and nonprofits](https://scalewithsearch.com/for/churches-and-nonprofits) - [All industries](https://scalewithsearch.com/for/) - Learn - For your office - [Office job guides](https://scalewithsearch.com/guides/) - [Browser calculators](https://scalewithsearch.com/tools/) - Start here - [How it works](https://scalewithsearch.com/how-it-works) - [Free Starter Kit](https://scalewithsearch.com/kit/business-memory-starter-kit.zip) - [Synthetic specimen](https://scalewithsearch.com/specimen/working-session-specimen.zip) - Guides - [The Complete Guide to Business Memory for AI Agents](https://scalewithsearch.com/articles/business-memory-for-ai-agents-guide) - [The Complete Small-Business Guide to AI Agent Governance](https://scalewithsearch.com/articles/ai-agent-governance-guide-small-business) - [The Complete Guide to Leaving Vendor AI Memory](https://scalewithsearch.com/articles/leaving-vendor-ai-memory-guide) - Articles by cluster - [Business memory](https://scalewithsearch.com/articles/business-memory-for-ai-agents-guide) - [Agent governance](https://scalewithsearch.com/articles/ai-agent-governance-guide-small-business) - [Migration and ownership](https://scalewithsearch.com/articles/leaving-vendor-ai-memory-guide) - For machines - [llms.txt](https://scalewithsearch.com/llms.txt) - [llms-full.txt](https://scalewithsearch.com/llms-full.txt) - [Machine view](https://scalewithsearch.com/?view=machine) - Company - Evidence - [Proof](https://scalewithsearch.com/proof) - Company - [About](https://scalewithsearch.com/about) # Keep hiring standards and candidate notes in files so AI works from evidence. You ask an AI assistant for a senior developer job description. It returns generic requirements that do not match your stack or your standards. The next week you ask for another posting, and the assistant remembers nothing. You explain your hiring criteria, pay ranges, required certifications, and interview process again. A recruiter runs several open roles at once. Each role has its own requirements, each candidate needs an individual evaluation, and each hiring manager has different priorities. Without memory, every request to the assistant starts from nothing. Recruiting memory is a set of files that hold your standards, your open roles, and your candidate notes between sessions. The assistant reads them before each task. The roles and candidates below are fictional. ## See why the chat cannot hold recruiting context Recruiting produces reference material every day: job descriptions, phone screen notes, interview scorecards, pay data by role and level, and skill lists by department. You cannot paste all of it into every conversation. The context window fills, so you start to cut. The cuts remove what matters: a candidate's specific technical background, the exact reason an applicant did not advance, or the precise certification a compliance role needs. Standard chat sessions also reset. The conversation about the engineering hire does not inform the one about the sales role. Tuesday's interview notes are gone by Thursday. ## Set up three kinds of file **Hiring standards.** One file, `recruiting-standards.md`, holds the rules that apply to every hire: - required qualifications by role type; - the standard interview process; - pay ranges by position and level; - technical assessment criteria and the scoring rubric; - the standard posting format, benefits text, and required legal text; - warning signs you have learned to check, each tied to job-related evidence. **Open roles.** One file per open position holds the full job description, the required skills, the hiring manager's notes, and the pipeline status. When you change a posting or close a role, the change stays in the file. **Candidate evaluations.** One file per candidate holds screen notes, assessment results, and each interviewer's feedback. Record why the candidate advanced or did not. The reason must be job-related and specific. ## Draft a posting from your standards You are opening a project manager role on the operations team. You ask: "Draft a job posting for the operations PM." The assistant reads the standards file and learns four rules for this company: 1. Manager roles require PMP certification. 2. Every posting lists the pay range. 3. The posting uses the standard benefits package and application steps. 4. The operations team needs vendor management and process documentation experience. The draft uses your format. The qualifications match your standards, and the responsibilities match the team's actual work. The pay range comes from your pay structure, and the legal text matches your state's requirements. Several US states now require a pay range in job postings. In those states, rule 2 is a legal duty, not only a preference. Check your state and city. You then add the role's own details: the vendor relationships the hire will manage and the software the team uses. Save those edits in the role file. The next operations PM posting starts from both the template and your changes. ## Capture screen notes while they are fresh On a busy day you screen six candidates. Each call shows something different: technical background, pay expectations, availability, and experience with your team's tools. Write the candidate file right after each call. Record technical skills, concrete projects that relate to the role, concerns the candidate raised, and questions the candidate asked. Write what the candidate said and did, not your impression of the person. Two weeks later, you choose who comes back for a second interview. You ask: "Compare the three senior developer candidates on React experience and team collaboration." The assistant reads the three files: | Candidate | React experience | Collaboration evidence | |---|---|---| | CAND-101 | 3 years | All examples are solo projects | | CAND-102 | 2 years | Specific examples of mentoring junior developers | | CAND-103 | 5 years | Screen answers showed limited depth on the questions asked | The comparison is specific because the notes were written when the evidence was fresh. The assistant compares; the hiring manager decides. ## Prepare the panel and score against one rubric Each panel member covers a different area: technical skill, working style, management ability, and project experience. Each question should draw out evidence, not a rehearsed answer. You ask the assistant for questions for each panel member, based on the role file and the candidate file. The role needs database tuning experience. The candidate claims PostgreSQL expertise. The assistant writes questions that require real knowledge: a query the candidate tuned, a production problem the candidate solved, and the trade-offs between two index types. After the interviews, each panel member adds feedback to the candidate file. The feedback scores skills against the standard rubric, quotes examples from the answers, notes concerns, and lists follow-up questions for the next round. For the final decision, you ask for a summary of the panel's view. The assistant combines the feedback, shows where the panel agrees and disagrees, and flags concerns that more than one interviewer raised. The summary is accurate because the notes are structured. It does not have to rebuild the interviews from memory. [AI memory for interview transcripts](/articles/ai-memory-interview-transcripts-summaries) shows how to keep the evidence, the summary, and the decision in separate records. ## Add patterns to the standards only after a test Over several hiring rounds, you notice patterns. Candidates who give specific examples of how they handled technical debt tend to perform well. Candidates who cannot explain the decisions on their past projects often struggle. Patterns like these are job-related, and they can go into the standards file. The assistant then uses them in future panels. The useful questions get asked more often, and new recruiters inherit what the team learned. Some patterns must not go in. "Candidates from certain bootcamps need more support" is a claim about a group, not evidence about a person. A pattern based on school, background, or "culture fit" can act as a proxy for a protected characteristic. Before a pattern enters the standards, confirm that it measures a job-related skill, and have a second person review it. When you fill a role again, the assistant can read the notes on past successful hires. It sees which skills mattered and which only looked good on paper. ## Run the setup The setup has three parts: 1. Claude Code, which runs in a terminal and reads files in a project folder. 2. A folder, for example an Obsidian vault, with the standards, role, and candidate files. 3. A `CLAUDE.md` file that tells the assistant where each record lives. No database or API work is required. The files sync through your normal cloud storage. The memory persists because it lives in files, not in the chat. Close the session today, open a new one tomorrow, and ask about last week's candidate. The assistant reads the same notes. ## Protect candidate data and keep the decision human Candidate files hold personal data. Collect what the hiring decision needs, limit who can read the folder, and delete files on the schedule your records obligations require. The [AI memory retention and deletion policy](/articles/ai-memory-retention-and-deletion-policy) gives a structure for that schedule. [What client data belongs in AI agent memory](/articles/client-data-in-ai-agent-memory) applies the same limits to any personal record. Laws on AI in hiring differ by place. In New York City, Local Law 144 requires a bias audit and candidate notice for automated employment decision tools. The city has enforced it since July 5, 2023. Check the rules where you hire before an AI output scores, ranks, or screens out a candidate. In the workflow above, the assistant drafts postings, writes questions, and summarizes notes. People screen, score, and decide. The assistant does not reject a candidate, send an offer, or contact an applicant. The [HR policy work comparison](/articles/claude-vs-chatgpt-hr-policy-work) shows how to test an assistant on a people-related job before you rely on it. ## Related: AI memory - [AI Correction Log Template That Changes the Next Run](/articles/ai-correction-log-template) - [CLAUDE.md Template for Business Context: What Claude Code Should Read Before Work](/articles/claude-md-template-business-context) - [How Corrections Become Part of the System](/articles/corrections-become-system-memory) ---- ```text .|########||. .|########||. .|########||. |##||. .||##|. |##||. .||##|. |##||. .||##|. |#|. .|#|. |#|. .|#|. |#|. .|#|. |#| |#| |#| |#| |#| |#| .#| |#. .#| |#. .#| |#. .#. .#| .#. .#| .#. .#| |#. .#| |#. .#| |#. .#| |# ...... #| |# ...... #| |# ...... #| .# ||#########| #| .# ||#########| #| .# ||#########| #| .#. |######||######|. .#. .#. |######||######|. .#. .#. |######||######|. .#. #. .##|###|##|#|######| .# #. .##|###|##|#|######| .# #. .##|###|##|#|######| .# || |##|#||||||||||||#||#| || || |##|#||||||||||||#||#| || || |##|#||||||||||||#||#| || # |#||||||||||||||||||||#| #. # |#||||||||||||||||||||#| #. # |#||||||||||||||||||||#| #. || |#||||||||||||||||||||||#| || || |#||||||||||||||||||||||#| || || |#||||||||||||||||||||||#| || # .#||||||||||||||||||||||||#| # # .#||||||||||||||||||||||||#| # # .#||||||||||||||||||||||||#| # || ....|||#||||||##|#|||#|#||##||||....|. || || ....|||#||||||##|#|||#|#||##||||....|. || || ....|||#||||||##|#|||#|#||##||||....|. || #. . ....|# ....#||| ||| .#||| ....#. .# #. . ....|# ....#||| ||| .#||| ....#. .# #. . ....|# ....#||| ||| .#||| ....#. .# # . ||||#| .#####|| . .#| .#||| ||||# #. # . ||||#| .#####|| . .#| .#||| ||||# #. # . ||||#| .#####|| . .#| .#||| ||||# #. .| ||||| #. |#|||||. || #. |#||. .||||| || .| ||||| #. |#|||||. || #. |#||. .||||| || .| ||||| #. |#|||||. || #. |#||. .||||| || || |.... #. ....|#. |. ...|. ....|| || || |.... #. ....|#. |. ...|. ....|| || || |.... #. ....|#. |. ...|. ....|| || #. .||||||##||||||##||####|||||||#||||||#| .# #. .||||||##||||||##||####|||||||#||||||#| .# #. .||||||##||||||##||####|||||||#||||||#| .# # .||####|###########################|. # # .||####|###########################|. # # .||####|###########################|. # # .#||||||.#.|| # |. ..# #| #|||||#| # # .#||||||.#.|| # |. ..# #| #|||||#| # # .#||||||.#.|| # |. ..# #| #|||||#| # # .#|||||| .. |# ## |#| ...#|#|||#| # # .#|||||| .. |# ## |#| ...#|#|||#| # # .#|||||| .. |# ## |#| ...#|#|||#| # # .#|||#|# .# .#| #| ##|.#..#|||||#| # # .#|||#|# .# .#| #| ##|.#..#|||||#| # # .#|||#|# .# .#| #| ##|.#..#|||||#| # # .#||||||############|######|||||#| # # .#||||||############|######|||||#| # # .#||||||############|######|||||#| # # |...||#||||||#||||#|#||||||#|||||#||| |.| # # |...||#||||||#||||#|#||||||#|||||#||| |.| # # |...||#||||||#||||#|#||||||#|||||#||| |.| # #. .. ||||# .|||##|. |#| .|| || .|||#. #| # .# #. .. ||||# .|||##|. |#| .|| || .|||#. #| # .# #. .. ||||# .|||##|. |#| .|| || .|||#. #| # .# || | ...|| ...##| | #| .|. |. ##### .. .| || || | ...|| ...##| | #| .|. |. ##### .. .| || || | ...|| ...##| | #| .|. |. ##### .. .| || || ||||| || ||||#| . |. |. |#. ||||| |#| || || || ||||| || ||||#| . |. |. |#. ||||| |#| || || || ||||| || ||||#| . |. |. |#. ||||| |#| || || .# |.....#|....|#.||||.|||##.|#|....||.#||.#. #. .# |.....#|....|#.||||.|||##.|#|....||.#||.#. #. .# |.....#|....|#.||||.|||##.|#|....||.#||.#. #. #.|||||||#############################| ||| .# #.|||||||#############################| ||| .# #.|||||||#############################| ||| .# || ##||#||#||||||#||#||#||#||##. || || ##||#||#||||||#||#||#||#||##. || || ##||#||#||||||#||#||#||#||##. || # .#|||||||||#||#|||||||||||#| # # .#|||||||||#||#|||||||||||#| # # .#|||||||||#||#|||||||||||#| # || |#||||||||||||||||||||||#| || || |#||||||||||||||||||||||#| || || |#||||||||||||||||||||||#| || .# |#||||||||||||||||||||#| #. .# |#||||||||||||||||||||#| #. .# |#||||||||||||||||||||#| #. || |#||||||||||||||||||#| || || |#||||||||||||||||||#| || || |#||||||||||||||||||#| || #. .######|#|#########| .# #. .######|#|#########| .# #. .######|#|#########| .# .#. |#####||||#####| .#. .#. |#####||||#####| .#. .#. |#####||||#####| .#. |# ||########|| #. |# ||########|| #. |# ||########|| #. |# ...... #| |# ...... #| |# ...... #| |#. .#| |#. .#| |#. .#| |#. .#. |#. .#. |#. .#. .#| |#. .#| |#. .#| |#. |#| |#| |#| |#| |#| |#| .|#|. .|#|. .|#|. .|#|. .|#|. .|#|. |##||. .||##| |##||. .||##| |##||. .||##| .||########||. .||########||. .||########||. Scale With Search 2026 [scalewithsearch.com](https://scalewithsearch.com) ```