Content Model Specialist (language modeling)

Tarrytown, NY, US • Posted 4 hours ago • Updated 4 hours ago
Full Time
On-site
Company Branding Image
Fitment

Dice Job Match Score™

🔢 Crunching numbers...

Job Details

Skills

  • Language Models
  • Analytics
  • Finance
  • Testing
  • Linguistics
  • Computational Linguistics
  • Science
  • Legal Research
  • English
  • Evaluation
  • Scripting
  • Virtual Team
  • Extraction
  • Taxonomy
  • Search Engineering
  • Regular Expression
  • Writing
  • Law
  • Machine Learning (ML)
  • Legal
  • Research
  • Privacy
  • Artificial Intelligence
  • Recruiting
  • MEAN Stack
  • Customer Service
  • Training And Development
  • SAP BASIS

Summary

Software Guidance & Assistance, Inc., (SGA), is searching for a Content Model Specialist for a full time employee position with our client, a content analytics firm in New York, NY. They turn contracts into data so that businesses can unlock the value of their contracts and take control of risk.
The team is comprised of legal, financial and technology experts with a passion for solving the complexity of contract analysis.

Responsibilities :
Our content models are the concept definitions, prompts, and rules that decide what the platform finds in contracts and how accurate it is. You will help define, test, and improve them. You'll work with our legal experts and our engineering team, turning legal judgment into precise, testable definitions and measuring how well they perform. You don't need to be a lawyer. You do need to be rigorous with language and quick to learn a new domain.
  • Work with legal experts to turn their judgment into operational definitions: inclusion and exclusion rules, examples, and decision rules for borderline cases.
  • Build and maintain a living guideline (codebook) that records decisions and their reasons.
  • Draft, test, and refine LLM prompts and search rules, and keep them consistent with each definition.
  • Build representative test sets from real contract text, including positive, negative, and borderline examples.
  • Analyze errors, find the cause (definition gap, ambiguous text, rule too broad, prompt issue), and fix the cause rather than the symptom.
  • Measure how well concepts perform and check that changes don't cause regressions.
  • Take part in regular reviews with legal experts on borderline cases, bringing options and evidence.
  • Work with Engineering on tooling and releases.
  • Help build a repeatable process for defining, testing, and releasing concepts.
Requirements:
  • Graduate degree (PhD, near-completion PhD, or master's) in linguistics, computational linguistics, a social science, or empirical legal research, or a closely related field.
  • Research or applied experience designing codebooks, annotation schemes, or similar operational definitions, ideally with checks that different people apply them consistently.
  • Strong written English and close-reading skills.
  • Skill at questioning domain experts and politely challenging vague or inconsistent answers.
  • A track record of getting up to speed in an unfamiliar domain.
  • Comfort with quantitative evaluation and error analysis.
  • Comfort working with structured data and a willingness to learn technical tools. Some scripting experience is helpful.
  • Ability to deliver practical results on product timelines.
  • Comfort working independently and communicating well in a fully remote team.
Nice to have:
  • Experience with information extraction, taxonomy, or search relevance.
  • Experience writing prompts as specifications.
  • Regex or rule-based pattern writing.
  • Exposure to legal or contract text.
  • Published or documented work on text analysis or annotation.
Not required:
A law degree, contract review experience, or machine learning engineering skills.
What success looks like:
  • 3 months: you understand contract language and our concepts, and have contributed several well-defined concepts with test sets.
  • 6 months: you work independently with legal experts and your concepts meet agreed accuracy thresholds.
  • 12 months: you contribute to a growing set of concepts and a documented process that others can use.
Hiring process
Portfolio and research review (thesis, papers, codebooks, or guidelines).
Clause-sorting exercise: write a definition from 10-15 real clauses, sort them into in, out, and borderline, and explain the borderline calls.
Simulation: given a vague request from a lawyer, show the questions you would ask.
Error analysis case: categorize the causes in 20 model outputs and propose fixes.
Conversation with our content experts and Engineering.

By applying for a job with SGA, you agree to allow SGA to process your application for this and future opportunities in accordance with our Privacy Policy. Also, to ensure timely processing, you agree to be contacted by our AI recruiter via email, text, or phone. Message frequency varies and data rates may apply, but you can reply STOP to any SMS message to opt-out of texts and may contact SGA at to opt-out of AI communications. The choice not to engage with AI will not adversely impact your consideration for placement. AI is not used to make any hiring determinations.

SGA is a technology and resource solutions provider driven to stand out. We are a women-owned business. Our mission: to solve big IT problems with a more personal, boutique approach. Each year, we match consultants like you to more than 1,000 engagements. When we say let's work better together, we mean it. You'll join a diverse team built on these core values: customer service, employee development, and quality and integrity in everything we do. Be yourself, love what you do and find your passion at work. Please find us at .

SGA is an Equal Opportunity Employer and does not discriminate on the basis of Race, Color, Sex, Sexual Orientation, Gender Identity, Religion, National Origin, Disability, Veteran Status, Age, Marital Status, Pregnancy, Genetic Information, or Other Legally Protected Status. We are committed to providing access, equal opportunity, and reasonable accommodation for individuals with disabilities in employment, and our services, programs, and activities. Please visit our company to request an accommodation or assistance regarding our policy.
Employers have access to artificial intelligence language tools (“AI”) that help generate and enhance job descriptions and AI may have been used to create this description. The position description has been reviewed for accuracy and Dice believes it to correctly reflect the job opportunity.
  • Dice Id: sgainc
  • Position Id: 26-02746
  • Posted 4 hours ago

Company Info

About Software Guidance & Assistance

Founded in 1981, SGA is a technology and resource solutions provider with a national footprint and headquartered in the shadow of Wall Street. We’re a certified women-owned business. We provide contingent staffing, direct placement, and professional and managed services to transform businesses and evolve careers. We’re small enough to tailor our services to each client and big enough to deliver for some of the world’s largest employers. Our professionals are experts in areas such as IT, finance, accounting, risk, and clinical.

SGA provides contingent staffing, direct placement, and professional and managed services nationwide for Fortune 500 companies, mid-size businesses and select startups.

Our core skillsets include all areas of technology – business & data analysis, cyber & network security, database administration, development & architecture, infrastructure, program & project management, quality assurance & testing. We also deliver talent across professional business functions such as finance, accounting, risk, and clinical.

Our Professional & Managed Services team delivers IT projects through onshore, offshore and hybrid delivery models. We develop software products, modernize applications, add features, and integrate and maintain systems. Our scope covers, among others, complex application suites, data management and visualizations, machine learning and mobile applications.

About_Company_OneAbout_Company_Two
Create job alert
Never miss an opportunity! Create an alert based on the job you applied for.