Data Analyst – Modeling & Insights (Founding Team)

Affluense.ai | Bengaluru (In-office) | Full-time | 1–5 years experience
ABOUT AFFLUENSE.AI
Affluense.ai is building India's wealth intelligence layer. We help wealth managers, private bankers and family offices find, understand and engage High Net Worth Individuals (HNIs).
We're live with enterprise customers, including some of India's largest private wealth firms, and backed by Zeropearl VC, with angels Kunal Shah (CRED) and Pravin Jadhav (Dhan).
WHAT YOU'LL WORK ON
- Mine the data for signal — go deep into filings, shareholding changes, directorships, funding rounds, property and deal data. Find the patterns that separate a true HNI from a paper director, and prove they're real, not noise.
- Build and own net worth and wealth scoring models — estimate wealth from indirect evidence (a director's stakes, company financials, promoter holdings, professional income proxies) and calibrate them so a score means something.
- Build propensity and ranking models — predict who is likely to have a liquidity event, move money or respond to outreach, so RMs spend time on the right 50 people, not the wrong 5,000.
- Segment and cluster — find natural cohorts (new-age founders, legacy business families, professionals, ESOP-rich employees) and what drives wealth in each.
- Clean and link messy data — match the same person or company across sources
- Measure and validate — backtests, holdouts, error analysis. Tell us honestly where a model is wrong and by how much.
- Turn findings into product and sales insight — present what the data says to founders, and to customers when it matters.
WHAT WE'RE LOOKING FOR
- 1–5 years in data analytics or analytical data science, with at least one model you built that ran in production or drove real business decisions
- Strong Python (pandas, NumPy, scikit-learn; XGBoost/LightGBM a plus)
- Solid statistics — distributions, hypothesis testing, sampling bias, leakage, calibration. You know when a pattern is real.
- Hands-on with regression, classification, tree-based models and clustering, and knowing which to pick and why
- Comfort with messy, incomplete, contradictory data — you enjoy the detective work
- Structured problem-solving: given "how wealthy is this doctor?", you can frame it, pick proxies, estimate, and defend the number
- Clear communication with non-technical people
NICE TO HAVE
- Experience in credit risk, fraud, underwriting, alternate-data scoring, marketing propensity or customer segmentation
- Background at an analytics firm (Mu Sigma, Tredence, Fractal, LatentView, etc.) or a BFSI/fintech data team
- Familiarity with Indian financial and corporate data — MCA filings, shareholding patterns, capital markets
THIS PROBABLY ISN'T FOR YOU IF
- You mainly want to work on LLMs, RAG pipelines or GenAI demos
- You prefer dashboards and reporting over building models
- You need clean, labelled data handed to you before you can start
WHAT YOU GET
- Ownership of the models that are the product — not a ticket queue on someone else's roadmap
- Direct mentorship from a founder who has built and scaled data science teams
- Competitive salary plus meaningful ESOPs in a funded, fast-moving team

Work arrangement
No