КазахстанData Science, Product Analyst
1 мес. назад
КазахстанOnsite
data analysissqlproduct metricsa/b testing
Responsible for owning and defining product metrics and measurement at Higgsfield AI to inform fast product decisions in generative AI video creation.
О компании
- Higgsfield AI is the fastest-scaling generative AI company in history, hitting $500M in annual revenue run rate, 25M+ users worldwide, 6M+ generations per day, and powering 390 of Fortune 500 brands. We're building at the absolute frontier of AI-powered video creation and next-generation creative tools. Joining Higgsfield means becoming part of a high-impact team shaping the future of AI-native experiences, at a company that isn't just moving fast, but rewriting what fast looks like.
Условия
- Competitive base salary in USD , based on your experience, skills, and the scope of the role.
- Equity participation through the company’s stock option program, giving you the opportunity to share in Higgsfield’s long-term growth.
- Relocation support to Almaty for candidates moving from another city or country.
- A highly collaborative, fast-paced environment where you can work directly with experienced leaders and have a meaningful impact on the product and company.
- Opportunities for professional growth, ownership, and career development as the company scales.
- Company-provided equipment, meals, transportation, or other office benefits.
- This is a fully on-site role based in our Almaty office . Our team works from the office five days per week for the full working day . We believe in-person collaboration is an important part of how we move quickly, solve complex problems, and build strong teams.
Другое
- This role owns how Higgsfield measures its product - from the first sign-up to whether people come back.
- You are the measurement owner for onboarding, first generation, the paywall, plans and credits, and retention. Product managers ship fast here - your job is to make sure they ship knowing.
- You make sure:
- Every product decision has a number attached to it before it ships and after it ships
- Experiments are designed to be readable, and then read honestly - including when the answer is “this didn’t work”
- The events the product emits can be trusted; you own the definition, not just the query
- A PM gets an answer in hours, not next sprint
- You are the person who decides what “it worked” means.
- Own the core product metrics: activation, first-generation success, generation depth, free→paid conversion, repeat usage, retention by cohort and segment.
- Define each metric once, write the definition down, and hold the line on it - one definition across dashboards, decks, and Slack.
- Instrument new features before launch, with product and engineering: which events, which properties, what success looks like, when we call it.
- Watch the health of the event stream itself - tracking drift, double counting, missing parameters, naming breakage and find the problem before a decision gets made on top of it.
- Design A/B tests properly: hypothesis, one primary metric, unit of randomization, MDE, sample size, run length, guardrail metrics.
- Read them honestly: significance, peeking, novelty effects, sample-ratio mismatch, segment heterogeneity.
- Separate “we proved this works,” “we proved this doesn’t,” and “we can’t tell from this test” - and say which one out loud.
- In-product funnel work: onboarding and quiz, first generation, paywall, checkout, plan choice, credit top-ups.
- Pricing and packaging analysis: plan mix, credit consumption, unit economics per generation, margin by feature and by model.
- Feature adoption and its real effect on retention and revenue — separating “users who do X retain better” from “making people do X improves retention.”
- Behavioural segmentation: casual creators vs corporate/B2B, new vs returning, by model and by use case. Mixing them hides everything that matters.
- Build the few dashboards PMs actually open on their own — decision-shaped, not comprehensive.
- Write short readouts a non-analyst can act on. Visualize data for humans, not for other analysts.
- Automate recurring reporting so your hours go to new questions, not refreshes.
- Work in SQL, Python, BigQuery and product analytics tools. We use AI heavily - resourcefulness beats syntax.
- We're looking for an analyst with opinions.
- You should have:
- Strong SQL: window functions, cohorts, funnels and retention on raw event data, without help.
- Python at working level for analysis (pandas, notebooks). Entry level is fine - we use AI heavily.
- Real experimentation experience: you have designed tests, run them, and killed features with the results.
- Statistical honesty: you know what a p-value does and does not entitle you to say, and you have refused to call a flat test a win.
- Understanding of subscription + one-time purchase mechanics: recurring vs one-off revenue, refunds, plan changes, deferred value of unspent credits.
- The instinct to check the instrument before explaining the movement - a surprising number is a claim, not a fact.
- Ownership: you decide what's worth analyzing, you chase the fix, you follow the recommendation to a shipped change.
- Clear written and spoken English, B2+.
- Product analysts from consumer subscription or PLG products
- Growth/BI analysts who moved into product and stayed
- Early-startup analysts who built product measurement from zero
- Data scientists who got tired of models and want decisions
- This role is not a fit if you:
- Wait for a ticket to tell you what to analyze
- Want to build models more than you want to change decisions
- Need a data engineering team and clean tables to exist before you can start
- Would report a lift you don't believe in because a stakeholder wants it
- Need perfect data before you can say anything useful
- Want predictable 9–5 workdays