Curacity - Sr. Data Analyst

Connecticut Innovations, Inc.; New York (NY); Hybrid; USD 110,000 - 140,000; Full time

Job summary

Curacity is seeking a results-oriented analytics professional to measure how travel properties appear in AI answers and to connect AI visibility to client website performance. You will own metrics, assess ranking and description quality, and work across engines and markets.

You will design measurement experiments, maintain dashboards, and report findings to leadership. Proficiency in SQL and Python is required, with Databricks experience valued; familiarity with AI tools and content performance

Qualifications

• 6+ years in analytics, data science, research, or search and content performance measurement.
• 2+ years on AI search, SEO / GEO, content performance, or LLM evaluation.
• Strong SQL and Python; Databricks experience preferred.

Responsibilities

• Measure AI visibility and outcomes across engines and markets.
• Design experiments and establish comparison groups; report changes over time.
• Evaluate content performance and advise product strategy.

Skills

Analytics experience

AI search experience

SQL

Python

Tools

Databricks

Job description

Are you ready to join Connecticut Innovation’s vibrant community of innovators? Connecticut Innovations (“CI”) is Connecticut’s strategic venture capital arm, and we are passionate about serving our portfolio of 220+ companies across various industries, with strengths in life sciences, technology, and climate tech.

Come join Curacity: Hotel marketing software that turns media into revenue!!

Location: New York City (Hybrid) or Atlanta, GA

Reports to: VP of Product

About Curacity

Curacity is a two-sided marketplace connecting travel brands that need to be discovered with the publishers, travel writers, and content creators travelers actually trust. Hotels and cruise brands define the editorial coverage they need; our network of publishers and creators produces original content; Curacity coordinates the fulfillment and measures the outcomes.

We operate two products on that network. Newsletter places original editorial recommendations in front of publisher subscriber audiences and ties the resulting bookings back to the content that drove them. Vista AI builds the distributed editorial footprint that determines whether a property gets recommended when a traveler asks ChatGPT, Perplexity, Google AI Mode, or Gemini where to stay.

Curacity partners with publications including AFAR and Travel + Leisure, and was named among Inc. 5000's Top 20 Travel & Hospitality Companies and Digiday's Best Content Marketing Platform of 2024. We are headquartered in New York with an office in Stamford, CT.

Role Overview

Travelers increasingly decide where to stay by asking an AI system. Vista AI exists to make sure our clients' properties are among the answers — and to prove, credibly, that our work is why.

This role owns that proof. You will measure how often and how favorably clients appear across the major AI search and chat systems, trace which published content the models actually draw on, and connect movement in AI visibility to what happens downstream on client websites. You will also read a great deal of AI-generated text closely and judge its quality, because the numbers only mean something if someone has looked carefully at what the models are saying.

What makes this hard is that the thing being measured keeps moving. Ask the same question twice and you may get different answers. Providers update their models without notice, and a client's visibility can rise or fall for reasons that have nothing to do with anything we did. So the core skill is not running the measurement — it is designing measurement that can tell a real change from an ordinary fluctuation, and being willing to report that a result is inconclusive when it is.

We are a small team, and we expect this person to multiply themselves with tooling rather than absorb more hours. Fluency with AI — agents, reusable skills, LLM-assisted analysis — is a core qualification here, not a bonus line at the bottom of the posting. Writing good SQL is table stakes; what distinguishes a strong candidate is having already used these tools to do analytical work faster and at higher volume than a person could unaided, and having the judgment to know which outputs need verification before anyone acts on them.

1.MEASURE AI VISIBILITY & CHANGE

Own the metrics that define how clients show up in AI answers: how often they appear, how they are ranked and described, how they compare to a defined competitive set, and how all of that moves over time.

• Design measurement that can distinguish a real change from ordinary variation — sampling standards, baselines, comparison groups, and stated uncertainty on every figure reported.
• Maintain the tracked-prompt panels that measurement runs against, across engines, topics, and markets.
• Report change over time as the primary output: what moved, by how much, in which direction, and whether the movement is attributable to anything we did.
• Trace which sources AI systems draw on for a given topic and market, and characterize the mix — client-owned sites, publishers and press, travel platforms, destination organizations, social, and the rest.
• Link published content to whether and how it later shows up as evidence in AI answers, and how long that takes and lasts.
• Identify where a client's evidence base is thin, and turn that into a specific view of what coverage is missing.
• Test what characteristics make content more likely to be drawn on: source, framing, format, specificity, recency, structure.

3. DOWNSTREAM IMPACT

• Connect AI visibility to downstream behavior: referral traffic from AI surfaces, on-site engagement, and booking or conversion activity on client websites.
• Establish what can honestly be claimed about revenue impact from AI recommendations, given the real limits of available referral and conversion data, working with the Newsletter analyst so both products use one standard for what counts as a defensible revenue claim.
• Build the instrumentation view: what is measurable today, what is not, and what would have to change for it to be — and be explicit internally about where the chain from recommendation to revenue is currently broken rather than papering over the gap.

4. ANSWER QUALITY & EVALUATION

• Evaluate AI answers about client properties and destinations for factual accuracy, relevance, coherence, completeness, and consistency.
• Identify and categorize failure modes — invented details, outdated information, confusion between similar properties, unfavorable or misleading characterizations — and track them over time by engine.
• Build and maintain the rubrics, benchmarks, and evaluation sets that make quality assessment repeatable, and manage rater agreement where more than one person is scoring.
• Design and iterate how we query these systems. What we ask, how we ask it, and how we structure prompts materially affects what we observe, and improving that is part of the job — as is applying the same evaluation discipline to any AI-assisted features Curacity builds.

5. RESEARCH, PRODUCT & COMMUNICATION

• Stay at the front edge of how AI search systems retrieve, weigh, and cite sources, and translate provider changes into what they mean for clients and for our editorial strategy.
• Validate claims through primary research rather than repeating what the AI-search commentary industry asserts.
• Define the metrics behind the Vista AI customer experience in partnership with Product and Engineering.
• Publish reproducible methodology. Any figure we put in front of a customer should be re-derivable from documented inputs by someone who was not in the room

Qualifications

• 6+ years in analytics, data science, research, or search and content performance measurement, including at least two years working directly on AI search, SEO / GEO, content performance, or LLM evaluation.
• Demonstrated experience measuring outcomes in a noisy, shifting environment, and the instinct to build a comparison group before drawing a conclusion.
• Working fluency in experiment design applied to observational data: comparison-group construction, difference-in-differences or matched comparison, and honest treatment of statistical power.
• Strong analytical judgment with unstructured text: designing rubrics, coding qualitative signals into structured data, and reasoning carefully about ambiguous cases.
• Deep hands-on fluency with LLMs and agentic tooling — prompt design, evaluation harnesses, agents that collect and process data at volume. Strong SQL and solid Python are assumed; we work in Databricks.
• Excellent written reasoning. Much of the output is explanation — of a method, a discrepancy, or why an answer is wrong.
• Intellectual honesty under commercial pressure: the willingness to say a result is not distinguishable from noise when a client, or a colleague, wants it to be.

Preferred Experience

• SEO, GEO / AEO, or content performance measurement, particularly where the measurement surface was owned by someone else.
• Running an annotation or human-evaluation program, including rubric design and rater agreement.
• Web analytics and referral-traffic analysis (GA4 or equivalent), and familiarity with AI-visibility tooling and vendor data.
• NLP for text analysis at scale — embeddings, classification, entity extraction — in service of measurement.
• Published or externally presented methodology work, in any venue.
• Travel, hospitality, or publishing domain knowledge

Why Curacity?

You'll join an award-winning, fast-growing team at the intersection of luxury travel, analytics, and modern media. We offer competitive compensation, comprehensive benefits, and the opportunity to shape the measurement and pricing infrastructure behind a high-growth hospitality technology company.

Curacity is proud to be an Equal Opportunity Employer. We celebrate diversity and are committed to creating an inclusive environment for all employees.