About this collaboration

I’m looking for a research collaborator to help build and evaluate an NLP system for a project targeting a top-tier NLP venue (ACL 2027 Findings). This is early-stage, hands-on research work, not a shadowing or teaching role.

This is an unpaid collaboration, not paid employment. In exchange, contributors who complete the agreed scope of work are included as co-authors on the resulting paper — not just an acknowledgment — and I cover all publication costs (article processing charges, etc.), so there’s no cost to you either way.

To be clear about what “co-author” means here: it’s tied to genuinely substantial, independently verifiable contribution — real ownership of specific components of the work, not a token task added to justify a byline. That protects both of us, and it’s also just how authorship is supposed to work.

What you’d work on

• Structured annotation and verification of an evaluation benchmark
• Building and testing an evaluation harness in Python
• Implementing baseline systems and running experiments
• Drafting evaluation scripts and result tables
• Literature and source-metadata collection

What you’d need

• Comfortable with Python and working with LLM APIs
• Careful, methodical annotation habits — this is a high-stakes domain where sloppy verification undermines the whole benchmark
• Able to work independently for stretches with a short weekly check-in, rather than needing close supervision
• Prior exposure to NLP/ML research is a strong plus, not a requirement

Logistics

• Remote, part-time — roughly 8–12 hours/week
• Runs through mid-December 2026
• One 30-minute check-in per week, async written updates otherwise
• Open until filled

How to apply

Email kaysarulanas2@gmail.com with a short note on your relevant experience and a CV/resume. Please put “Research Collaboration” in the subject line.