Install
$ agentstack add skill-harshuljain13-agent-skills-eb1a-judging ✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ✓ Network access No
- ✓ Filesystem access No
- ✓ Shell / process execution No
- ✓ Environment & secrets No
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
Verified badge
Passed review? Show it. Paste this badge into your README, it links to the public security report.
Reliability & compatibility
Declared compatibility
Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.
We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.
How agent discovery & health will work →About
EB1A Judging — Hackathon Outreach Skill
Context
Harshul is a Senior ML Engineer at Audible (Amazon), ~9 years experience. He judges hackathons for two reasons: he genuinely enjoys evaluating real engineering work, and every judging engagement is evidence for Criterion 4 (Judging) of his EB1A extraordinary ability petition.
Every outreach email, confirmation, and post-event letter gets documented.
Task 1: Write the Outreach Email
Harshul's email voice: direct opener with credentials, clear ask, bullet-pointed fit, warm close. No fluff, no over-selling.
Template
Subject: Volunteer Judge/Mentor — [Hackathon Name]
Hi [Name / "there"],
I'm a Senior ML Engineer at Audible (Amazon) with 9 years building production systems
and 6 years specifically in ML/AI. I'd like to volunteer as a judge and mentor
for [Hackathon Name].
Why I'm a fit:
- Production AI: I build GenAI-powered search and a real-time ML feature store
(100K+ QPS) at Audible — I evaluate production-readiness, not just demo quality
- Open source: I created AgentShip, an AI agent deployment framework
- Mentorship: 300+ engineers mentored on Codementor/Topmate (4.95/5, 298 sessions)
- What I look for: working systems over polished slides — technical depth, honest
tradeoffs, and what the team actually learned
Happy to review submissions, do real-time mentorship, or sit on a final panel —
whatever's most useful. Can we find 15 minutes to talk?
Best,
Harshul Jain
[LinkedIn] | [Email]
Customization — always do this:
- Theme-specific line: Add 1 sentence connecting his background to the hackathon's focus
- AI/ML theme → "Your focus on [LLMs / agents / computer vision] maps directly to what I build day-to-day."
- Social impact → "I'm especially interested in AI applied to [domain] — right in line with your theme."
- University / student event → "I love working with early-career engineers who are building real things."
- Named organizer: Always find the specific person's name if possible
- Event date: Include it if known — shows you've done your homework
Task 2: Post-Event — Request Documentation Letter
Send within 2 weeks of the event. This letter is EB1A evidence.
Subject: Quick Ask — Confirmation Letter for [Hackathon Name]
Hi [Name],
Thanks again for having me at [Hackathon Name] — I really enjoyed [specific moment
or project that stood out].
I'm compiling documentation for an immigration petition and would appreciate a
brief letter on [Organization] letterhead confirming my role. It only needs:
- My name and role (Judge / Technical Judge)
- Event name, date, format
- Rough scope (number of teams / submissions reviewed)
- Your name, title, org
Happy to send a draft if that's easier. Should take under 10 minutes.
Thank you,
Harshul Jain
[LinkedIn] | [Email]
Task 3: Log the Engagement (EB1A Evidence)
After each hackathon, create a log entry immediately. File it wherever Harshul keeps his EB1A evidence folder.
## Judging Log — [Hackathon Name]
Date: [Date]
Organizer: [Org + URL]
Format: In-person / Virtual / Hybrid
Scale: [N] teams, [N] participants
My role: Judge / Technical Judge / Mentor-Judge
How I was selected: [Invited / Applied / Referral]
What I evaluated: [e.g., "Reviewed 10 ML projects, scored on technical execution,
production-readiness, and novelty"]
Artifacts:
[ ] Invitation/confirmation email
[ ] Organizer confirmation letter (request within 2 weeks)
[ ] Screenshot of event page showing my name
[ ] Judging rubric I used
[ ] Any media or recap mentioning me
EB1A Criterion: 4 — Judging
Strength: High / Medium / Low
What makes it stronger: prestigious organizer (MLH, top university, AWS), large scale (100+ participants), evidence you were selected not just self-nominated, named in public event materials.
EB1A Framing (for petition narrative)
> "Petitioner was selected by [Org] to evaluate the work of [N] engineering teams at > [Hackathon Name], applying expertise in ML and production systems developed over > 9 years at companies including Amazon. This selection reflects recognition of > Petitioner's standing in the field by peers and community organizers."
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: harshuljain13
- Source: harshuljain13/agent-skills
- License: MIT
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet, be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.