D
Defcon AI
Defcon AI logistika, transport va ta’minot zanjiri operatsiyalari uchun sun’iy intellektga asoslangan dasturiy ta’minot ishlab chiqadi. Kompaniyaning texnologiyasi tashkilotlarga tabiiy ofatlar, kutilmagan hodisalar va murakkab sharoitlar sabab yuzaga keladigan uzilishlarni rejalashtirish va ularga javob berishga yordam berish uchun dasturiy modellashtirishni intellektual agentlar bilan birlashtiradi. Sun’iy intellekt, mobillik va logistika kesishmasida faoliyat yurituvchi kompaniya operatsion barqarorlikni, qaror qabul qilishni va javob berish samaradorligini oshirishga mo‘ljallangan ma’lumotlarga asoslangan vositalarni taqdim etish uchun hamkorlar bilan hamkorlik qiladi.

Senior Entity Resolution Engineer (Remote, United States)

Build production-grade entity resolution and graph provenance systems for DEFCON AI’s platform. Apply record matching, explainable decision-making, and data lineage techniques to support complex government missions.

Tavsif

  • Develop and improve entity-level matching workflows covering blocking, candidate generation, pairwise scoring, clustering, and threshold policies
  • Use a shared matching engine for record linkage, deduplication, and known-entity verification
  • Create provenance tracking that links every graph node and edge to its asserting source
  • Manage the tradeoff between false merges and false splits, making matching decisions clear and explainable
  • Deliver the matching implementation against the internal design, separately evaluate the reference dataset, candidate retrieval, and final matching, and provide evidence-based threshold recommendations for review
  • Document the matching methodology thoroughly enough to guide implementation by other engineers

Talablar

  • At least five years of experience, including production delivery of record matching or entity resolution systems
  • Ability to articulate matching tradeoffs, including approaches to false merges and false splits
  • Strong Python and SQL skills, with experience working on large, messy, real-world datasets
  • Ability to explain matching outcomes to stakeholders who need to defend decisions without knowing the underlying implementation
  • US citizenship is required
  • An active US Secret clearance is required at the start of employment
  • Preferred: hands-on probabilistic matching with inconsistent identity data, including names, dates, addresses, and identifiers, plus knowledge of their failure modes
  • Preferred: familiarity with probabilistic record-linkage frameworks and tools such as Fellegi-Sunter models, Splink, Dedupe, Zingg, or an equivalent internal system
  • Preferred: experience in record linkage, master data management, or identity management
  • Preferred: production experience with graph data modeling and graph algorithms
  • Preferred: PostgreSQL and pgvector or comparable technologies
  • Preferred: active Top Secret clearance

Imtiyozlar

  • Fully remote, results-oriented work environment
  • Competitive salary with bonus and equity
  • Employer-paid medical, dental, and vision coverage for employees and their families
  • Unlimited paid time off with manager approval
  • Flexible scheduling with control over the workday
  • Fourteen weeks of fully paid parental leave

O‘xshash ish o‘rinlari