Founder
Finding AI Talent on Hacker News — 5 Engineers Worth Reaching Out To
Key Takeaway
AI agents scan Hacker News for top AI and FHE talent — filtering "Who is hiring?" threads, prolific commenters, and technical depth signals — and produced a ranked shortlist of 5 engineers worth outreaching, including 3 FHE researchers we'd never have found otherwise.
The Problem
Hiring in AI is a disaster. Everyone's competing for the same pool of ML engineers who already have 6 offers. LinkedIn is pay-to-play. Recruiters send the same 50 candidates to every company.
The talent we actually want — FHE researchers, agent infrastructure engineers, people who think about multi-agent coordination — they're not on LinkedIn changing their status to "Open to Work." They're on Hacker News arguing about lattice cryptography at 2 AM.
The signal is there. People reveal their expertise through what they write about, argue about, and build. A Hacker News commenter who writes 500 words explaining why TFHE outperforms BGV for specific use cases — that person knows more about FHE than most PhD candidates. And they're self-identifying by contributing publicly.
The problem is finding them. HN has millions of comments. You can't read them all. You can't even search them effectively with HN's native tools.
The Solution
The Hacker News Scraper skill on Mr.Chief, combined with competitive intelligence analysis. The agent scans "Who is hiring?" threads, "Who wants to be hired?" threads, and prolific commenters on AI/FHE topics. Filters by technical depth, engagement quality, location compatibility, and relevant expertise. Outputs a ranked list with outreach angles.
The Process
The talent discovery pipeline:
yamlShow code
# Agent: Pauly — HN Talent Scout
skill: hacker-news-scraper
schedule: "0 10 1,15 * *" # Bi-monthly (1st and 15th)
scan_targets:
monthly_threads:
- "Ask HN: Who is hiring?"
- "Ask HN: Who wants to be hired?"
- "Ask HN: Freelancer? Seeking freelancer?"
topic_threads:
keywords:
- "homomorphic encryption"
- "FHE"
- "TFHE"
- "lattice cryptography"
- "privacy-preserving computation"
- "multi-agent systems"
- "agent orchestration"
- "AI infrastructure"
min_comment_score: 5
min_comment_length: 200 # Filters for depth
prolific_commenters:
topics: ["cryptography", "FHE", "AI agents", "Rust systems"]
min_comments: 10 # In last 6 months on relevant topics
min_avg_score: 3
filters:
location:
preferred: ["Paris", "France", "Europe", "Remote"]
acceptable: ["US", "UK", "Canada"]
expertise_signals:
strong: ["implemented", "built", "shipped", "contributed to", "published"]
medium: ["worked with", "experience in", "familiar with"]
weak: ["interested in", "learning about"]
output:
ranked_list: true
per_candidate:
- username
- expertise_summary
- top_3_comments # With links
- location_signals
- outreach_angle
- confidence_score
The agent's analysis process:
View details
Step 1: SCRAPE "Who is hiring?" for March 2026
→ 847 job posts, 234 "seeking work" comments
Step 2: FILTER for relevant expertise
→ 23 candidates mention AI/ML/crypto/privacy
Step 3: DEEP SCAN prolific commenters on FHE/agent topics
→ 156 unique usernames with 10+ comments on relevant threads
Step 4: SCORE each candidate
→ Technical depth (comment quality analysis)
→ Engagement pattern (consistent contributor vs drive-by)
→ Location compatibility (parsed from comments/profile)
→ Expertise match (FHE, agents, systems, Rust)
Step 5: RANK and produce shortlist
→ Top 5 with detailed profiles
Sample output:
markdownShow code
# HN Talent Scout — March 2026
## Top 5 Candidates
### 1. lattice_wizard (Confidence: 92%)
**Expertise:** FHE implementation, TFHE-rs contributor, Rust systems
**Location:** Lyon, France (mentioned in "Who wants to be hired?")
**Signal strength:** STRONG
**Top Comments:**
- [comment_link] 347-word explanation of TFHE parameter selection
with concrete benchmarks (42 points)
- [comment_link] Corrected a misconception about FHE noise growth
in a Zama thread (28 points)
- [comment_link] "I've been implementing a TFHE-based private
inference engine for the last 8 months" (19 points)
**Why this person:**
Active TFHE-rs contributor, clearly deep expertise, based in France,
and explicitly mentioned seeking new opportunities. Their comments
show practical FHE engineering, not just theoretical knowledge.
**Outreach angle:**
"Saw your TFHE parameter selection analysis on HN — we're the
team behind Zama and building agent infrastructure for FHE
applications. Your benchmarking approach matches exactly what
we need. Coffee in Lyon?"
---
### 2. agent_infra_dev (Confidence: 87%)
**Expertise:** Multi-agent systems, distributed systems, Python/Go
**Location:** Remote (Europe-based, timezone signals suggest CET)
**Signal strength:** STRONG
**Top Comments:**
- [comment_link] Detailed architecture post about agent-to-agent
communication patterns (67 points)
- [comment_link] "Built a 12-agent system for automated trading
— here's what breaks at scale" (34 points)
- [comment_link] Critique of CrewAI's memory implementation with
suggested improvements (22 points)
**Why this person:**
Hands-on multi-agent builder. Already thinks about production
scale issues. Actively critiques competitor frameworks —
would likely be interested in our approach.
**Outreach angle:**
"Your critique of CrewAI's memory system was spot on. We're
running 31 agents in production and solved exactly the problems
you described. Want to see how?"
---
### 3-5. [Additional candidates with similar detail...]
## Honorable Mentions (Candidates 6-10)
[Abbreviated profiles for secondary outreach]
## Cross-Reference Status
- LinkedIn enrichment: Requested (via LinkedIn skill)
- GitHub profiles: Found for 3/5 candidates
- Email: Not available from HN — use HN messaging or LinkedIn
The Results
| Metric | Traditional Recruiting | AI Agent Talent Scout |
|---|---|---|
| Talent pool scanned | LinkedIn applicants (self-selected) | HN's entire active community |
| Expertise verification | Resume claims | Public comment history |
| Time to shortlist | 2–3 weeks | 45 minutes |
| Cost | $5K–$20K (recruiter fees) | ~$0 |
| Quality signal | Interview-dependent | Pre-validated by community |
| Outreach personalization | Generic InMail | Comment-specific angles |
| Passive candidates found | Rarely | By design |
Outcomes from the first 3 months:
- 5 candidates identified per bi-monthly scan (30 total screened)
- 3 FHE researchers found through comment analysis — none were on LinkedIn as "looking"
- 1 hire made — an agent infrastructure engineer found through their CrewAI critique comments. They'd never applied anywhere. We found them through their opinions.
- 2 conversations initiated with FHE researchers who are now consulting part-time
- Response rate to HN-personalized outreach: 60% (vs ~5% for cold LinkedIn InMail)
The 60% response rate is the real story. When you open with "I read your comment about TFHE parameter selection and disagreed with one point — here's why" instead of "Hi, I found your profile interesting," people respond.
Try It Yourself
bashShow code
mrchief skills install hacker-news-scraper
Define your expertise keywords. Set the scan to run bi-monthly around "Who is hiring?" thread dates (first Monday of each month). The agent finds talent where they naturally congregate — not where they're being harvested by every recruiter on LinkedIn.
The best engineers don't have "Open to Work" on their LinkedIn. They have 500-point comments on Hacker News. That's where you find them — if you're paying attention.
Related case studies
Strategy Lead
Tracking Competitor Launches on Hacker News — We Knew Before Their Blog Post
How AI agents monitor competitor launches on Hacker News with sentiment analysis and feature comparison. Real-time competitive intelligence via Mr.Chief.
Founder
Daily Hacker News Digest — 10 Stories That Matter, Zero Noise
Automated Hacker News monitoring with AI agents on Mr.Chief. Get a daily curated digest of 10 relevant stories filtered for AI, FHE, crypto regulation, and dev tools.
Research Analyst
Academic Paper Alerts — New FHE Research Relevant to Zama's Roadmap
AI agents monitor arXiv, IACR ePrint, and conference proceedings for FHE research. Weekly digest curates 3 papers that matter from 50+ published. Powers Zama's R&D awareness.
Want results like these?
Start free with your own AI team. No credit card required.