Imported from apidojo-io/apidojo-skills (skills/intent/finding-data-scientists-on-twitter/SKILL.md). Install upstream with npx skills add apidojo-io/apidojo-skills --skill finding-data-scientists-on-twitter. Copyright stays with the author (Apache-2.0).

Finding Data Scientists And Ml Engineers on Twitter

Discovers data scientists and ML engineers on Twitter via skill keywords, portfolio/project signals, and open-to-work indicators. Twitter surfaces professionals who actively discuss their craft — a strong passive candidate signal.

Prerequisites

• APIFY_TOKEN environment variable set
• Optional: Apify MCP server installed

Inputs

Workflow

Progress:
- [ ] Step 1: Search for role-specific tweets
- [ ] Step 2: Collect unique handles
- [ ] Step 3: Enrich profiles
- [ ] Step 4: Score candidate fit
- [ ] Step 5: Deliver candidate list

Step 1: Search Queries

Recommended — run_actor.js (handles waiting, output, and file saving automatically):

# Quick answer (prints table to chat)
node scripts/run_actor.js \
  --actor "apidojo~twitter-user-scraper" \
  --input '{"param": "value"}'

# Save as CSV
node scripts/run_actor.js \
  --actor "apidojo~twitter-user-scraper" \
  --input '{"param": "value"}' \
  --output YYYY-MM-DD_results.csv --format csv

# Save as JSON
node scripts/run_actor.js \
  --actor "apidojo~twitter-user-scraper" \
  --input '{"param": "value"}' \
  --output YYYY-MM-DD_results.json --format json

APIFY_TOKEN must be set in environment or .env file.

If Apify MCP is available:

Tool: apify:run-actor
Actor: "apidojo~tweet-scraper"
Input:
{
  "searchTerms": ["data scientist", "ML engineer", "LLM engineer", "machine learning open to work"],
  "maxItems": 300,
  "tweetLanguage": "en"
}

REST API fallback:

curl -X POST \
  "https://api.apify.com/v2/acts/apidojo~tweet-scraper/runs?token=$APIFY_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"searchTerms": ["data scientist", "ML engineer", "LLM engineer", "machine learning open to work"], "maxItems": 300}'

Collect unique author.username from results.

Step 2: Enrich Profiles

If Apify MCP is available:

Tool: apify:run-actor
Actor: "apidojo~twitter-user-scraper"
Input: {"usernames": ["[username1]", "[username2]", "..."]}

REST API fallback:

curl -X POST \
  "https://api.apify.com/v2/acts/apidojo~twitter-user-scraper/runs?token=$APIFY_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"usernames": ["handle1", "handle2"]}'

Step 3: Filter and Score

Skill confirmation: bio contains keywords: "data science", "machine learning", "ML", "NLP", "LLM", "AI", "neural network", "PyTorch", "TensorFlow", "scikit-learn"

research_signal = bio contains 'PhD', 'researcher', or links to papers/Google Scholar

Candidate score:

candidate_score = (skill_confirmed ? 1 : 0) * 0.35
                + (open_to_work_signal ? 1 : 0) * 0.30
                + (followerCount in 200..20000 ? 1 : 0.6) * 0.20
                + (tweeted_in_last_30_days ? 1 : 0) * 0.15

Activity: active (< 30 days) | passive (30–90 days) | dormant (> 90 days)

Step 4: Edge Cases

• Company/brand accounts in results: Filter where followerCount > 50K AND bio contains no personal pronouns; these are likely brand accounts
• < 20 candidates found: Broaden skill term; remove location or seniority filter; try adjacent skills
• Bot detection: Flag followerCount / followingCount < 0.05 AND tweetsCount < 20 as potential bot
• Location not matching: Bio location is free text — use fuzzy match; accept partial city/country names

Output Format

# Data Scientists And Ml Engineers Candidates: [ML_SPECIALTY]
Profiles found: [N] | Open-to-work: [N] | Active: [N] | Date: [DATE]

## Priority: Open-to-Work Candidates
| Name | @Handle | Specialty | Location | Followers | Last Active | Score |
|------|---------|----------|---------|-----------|------------|-------|

## Passive Candidates
| Name | @Handle | Specialty | Location | Followers | Score |
|------|---------|----------|---------|-----------|-------|

## Bio Highlights (Top 5)
1. @[handle]: "[bio excerpt]"

Troubleshooting

All results are agencies/companies not individuals: Add personal pronouns filter or search "I am a [role]", "I do [skill]". Role too generic returns too many results: Add location OR seniority qualifier. No open-to-work signals: Most candidates don't signal publicly — treat passive candidates as warm leads with personalized outreach referencing their recent content.