Remote | Formal Verification & Lean 4 Specialist — Up to $90/hour

24-Mag • New York, NY

4d • $90 - $90 • Remote

About The Position

We are sharing a specialised part-time consulting opportunity for formal-methods professionals with hands-on expertise in Lean 4, interactive theorem proving, mathematical formalisation, program verification, and proof evaluation. This role supports an advanced AI research initiative focused on strengthening expert-level reasoning in frontier models. Selected professionals will develop challenging formal-verification and theorem-proving problems, review peer-authored tasks, and evaluate AI-generated proofs, tactics, and formalised mathematics for correctness, completeness, and technical rigour.

Requirements

• Hands-on expertise in Lean 4, interactive theorem proving, mathematical formalisation, program verification, and proof evaluation
• Substantial experience in formal verification or interactive theorem proving
• Hands-on proficiency with Lean 4 and mathlib
• Strong familiarity with type theory, mathematical logic, and formal semantics
• Experience formalising mathematical theorems or verifying software properties
• Ability to read, write, debug, and evaluate machine-checked proofs
• Excellent technical writing and meticulous attention to detail
• Experience reviewing advanced formal-methods or mathematical work
• An advanced degree in computer science, mathematics, logic, software engineering, or a related discipline is highly relevant
• Equivalent professional or research experience in theorem proving and formal verification may also be considered
• Academic research involving type theory, proof assistants, programming languages, or verification may strengthen an application
• Publications, formalisation projects, open-source contributions, or substantial proof-assistant work may also be valuable

Nice To Haves

• Deep experience with Lean 4 and the mathlib ecosystem
• Expertise in programme verification, dependent types, or proof automation
• Familiarity with Coq, Isabelle/HOL, Agda, HOL Light, or similar systems
• Experience formalising advanced mathematics
• Background in compiler verification, protocol verification, or verified systems
• Experience teaching, grading, peer reviewing, or authoring formal-methods problems
• Previous exposure to AI proof evaluation, benchmark development, or technical training-data projects
• Familiarity with automated theorem proving and proof-search systems
• Experience with Coq, Isabelle, Agda, or comparable proof assistants

Responsibilities

• Design expert-level problems involving theorem proving, program verification, and mathematical formalisation
• Create tasks requiring precise logical reasoning and valid machine-checkable proofs
• Develop clear problem statements, assumptions, definitions, and expected outcomes
• Prepare technically sound reference proofs and formal solutions
• Ensure tasks test genuine formal reasoning rather than superficial syntax generation
• Develop and review formal proofs using Lean 4 and mathlib
• Assess theorem statements, tactic sequences, term-style proofs, and supporting definitions
• Identify type errors, missing assumptions, invalid proof steps, and incomplete formalisation
• Evaluate whether proofs compile and satisfy the intended mathematical claim
• Review alternative valid proof strategies where appropriate
• Review problems and reference solutions authored by other formal-methods specialists
• Assess clarity, originality, difficulty, and technical correctness
• Verify that formal specifications accurately represent the intended theorem or programme property
• Identify ambiguity, unintended shortcuts, inconsistent assumptions, and grading gaps
• Recommend revisions that improve proof quality and evaluation reliability
• Review and compare AI-generated proofs, tactics, and formalisation attempts
• Assess logical validity, syntactic correctness, completeness, and alignment with task requirements
• Identify hallucinated lemmas, invalid imports, brittle tactics, and unsupported reasoning
• Assign Accept, Revise, or Reject outcomes with detailed written rationale
• Explain technical failures clearly and propose appropriate corrections

Benefits

• Flexible remote work
• Competitive hourly compensation
• Weekly payments via Stripe or Wise
• Projects may be extended, shortened, or adjusted depending on scope and performance

What This Job Offers

• Job Type: Part-time
• Career Level: Senior