Source passive candidates without LinkedIn Recruiter

Founder or first-in-house recruiter at a 20-100 person startup hiring for specialized roles (engineers, designers, senior ICs) where job posts alone don't attract enough quality candidates · Assemble in ~10 min; first personalized sequence in <3 min once installed.

HiringEdit flowon X● golden 89

Not legal or HR-compliance advice. Employment and hiring rules vary by jurisdiction — have qualified counsel review policies, offers, and screening before use.

step 1ready

Define the role and ideal candidate profile

hiring10 alts

step 2connector

Source candidate names and contact info

⚡ bring your own tool

step 3ready

Fan-out research per candidate

outreach-fanout-research31 alts

step 4partial

Score fit against role profile

prospecting

step 5ready

Write personalized recruiting outreach sequence

outreach9 alts

step 6ready

Add deliverability and cadence structure

outreach-expert6 alts

step 7connector

Launch and track via outreach tool

⚡ bring your own tool

step 8ready

Draft reply responses and schedule screens

cold-email-outreach6 alts

Did this recipe work for you?

What you get

▸Candidate profile brief — structured role spec and sourcing criteria
▸Per-candidate research notes — verified public signals per person
▸Ranked candidate list with fit scores
▸Personalized multi-touch outreach sequences (first message + 2 follow-ups per candidate) ready to load into your sender

tools: LinkedIn free / Sales Nav from $99/mo · Hunter.io from $0 (25 free/mo) · Instantly or Lemlist from $37/mo · Calendly free tier

tokens: ~$2-6 per 20-candidate batch (research fanout dominates)

≈ $40-140/mo at active hiring volume depending on sender tool

What stays yours

· Review and approve each message sequence before loading into sender — personalization errors here damage your employer brand
· Conduct the actual phone/video screens and make the hire decision
· Update the candidate tracker after each reply so follow-up timing stays accurate

✓ Last verified 2026-06-11claude-sonnet-4-6 (subagent, test-mode fixtures)

✓ step 1 · Define the role and ideal candidate profile

✓ step 2 · Source candidate names and contact info

✓ step 3 · Fan-out research per candidate

✓ step 4 · Score fit against role profile

✓ step 5 · Write personalized recruiting outreach sequence

✓ step 6 · Add deliverability and cadence structure

✓ step 7 · Launch and track via outreach tool

✓ step 8 · Draft reply responses and schedule screens

view sample outputs from this run

step 1 · Define the role and ideal candidate profile

# Candidate Profile Brief
**Role:** Senior AI/ML Infrastructure Engineer  
**Hiring Manager:** Mara Lin, CTO — Lattice AI (Series A, ~35 people)  
**Target Start:** Q3 2026 | **Compensation:** $200–240k base + 0.3–0.6% equity  

---

## Role Summary
Lattice AI is building a multi-tenant LLM serving platform for enterprise customers. We need a senior engineer who can own the inference serving stack — model routing, batching, latency/cost optimization — and help us scale from 50 to 500 enterprise tenants. This person will be IC #3 on the infra team, reporting to the CTO.

---

## Required Skills…

step 3 · Fan-out research per candidate

# Candidate Research Briefs
Generated: 2026-06-11 | Recipe: Source passive candidates without LinkedIn Recruiter

Each brief was assembled from public signals: GitHub profile, merged PRs, blog posts, conference recordings, and Twitter/X history. No private data used.

---

## Candidate 1 — Jordan Wei
**Title:** Staff Inference Engineer | **Company:** Together AI  
**Contact:** jordan.wei@protonmail.com | linkedin.com/in/jordanwei-ml | github.com/jordanwei  

### Work History
- Together AI (2023–present): Staff Inference Eng — owns continuous batching pipeline; reduced P99 TTFT by 38% (per publ…

step 4 · Score fit against role profile

# Ranked Candidate List
Scored against: candidate-profile-brief.md (Senior AI/ML Infrastructure Engineer, Lattice AI)
Scoring model: ICP-style fit scoring mapped to recruiting signals (see field notes per candidate)

---

## Scoring Rubric

| Dimension | Weight | Description |
|-----------|--------|-------------|
| Technical fit | 30% | LLM serving stack depth; CUDA/Triton/kernel experience; required tech match |
| Tier signal | 20% | OSS contributions, title at LLM provider (Tier 1/2/3 from profile brief) |
| Seniority match | 20% | IC level, years in production infra (not research) |
| Cultu…

The problem

When I open a hard-to-fill role, I want a list of qualified passive candidates with personalized outreach messages drafted and ready to send, so that I can start conversations without spending $8,000/year on LinkedIn Recruiter or 20 hours manually searching.

Cybernetic layers

Memorybyo

Candidate state (who was contacted, current stage, reply status) needs a lightweight tracker — Google Sheet, Airtable, or a local CSV updated after each batch. No harvested memory artifact covers recruiting pipeline state yet.

Backgroundbyo

Follow-up cadence timing (day 3, day 7 nudges) is manual today — the outreach tool handles scheduling once loaded. A Claude Code cron hook could auto-draft follow-ups but requires a memory layer to know who's pending.

Harnessready

Claude Code is the runtime. Drop each skill into ~/.claude/skills/ (or install via Cowork) and invoke by name. The recruit-config/user.md file the outreach skill reads must be created once with your name, email, and tone preferences.

Click any step to see what fills it — and the ranked alternatives you can swap in.