Data Scientist jobs in Delhi at CryptoMize are open on a rolling, always-hiring basis — we staff modeling capacity ahead of the engagements that demand it, not after. This is a full-time, permanent position with immediate joining at our New Delhi HQ, inside the forecasting core whose models have tracked real electoral and reputation outcomes at 89% historical accuracy across 18 countries. The complete job description follows: the responsibilities you will own, the requirements, the seniority path, and the selection process. Practicing data scientists who are tired of models that die in dashboards — and want theirs to face reality — should read to the end.
LOCATION New Delhi (HQ)
EMPLOYMENT Full-time · Permanent
AVAILABILITY Immediate · Rolling intake
COMPENSATION Discussed at screening
TRACKS ON THIS DESK6
CRAFT SKILLS NAMED12
TOOLS & SYSTEMS5
PATH STAGES4
01
01The actual work
What will you actually do as a Data Scientist at CryptoMize?
01
Own the modeling layer of live engagements end-to-end — problem framing with the strategist, model design, validation, and the sign-off on what the number actually means
02
Run forecasting at consequence: electoral trajectory, sentiment inflection, and risk-signal models whose outputs reach client counsel directly
03
Design segmentation and targeting models for perception and political engagements — the unsupervised work our platforms consume at scale
04
Build the experimental and quasi-experimental measurement behind reputation interventions, separating real movement from noise with statistical honesty
05
Set reproducibility standards for the cell — versioned models, pinned data, method notes a peer can reconstruct line by line
06
Mentor Data Scientist interns and analysts; the teaching culture is why the cell compounds
07
Carry the miss reviews: every forecast that missed gets a documented post-mortem, and the discipline behind our 89% record is that we study failures more closely than wins
ROLE RESPONSIBILITIES
As a Data Scientist at CryptoMize you will frame engagement problems into modeling problems — the hardest and most valued step in our pipeline — then carry them through data preparation, model selection, validation, and delivery. The role applies rigorous validation before any forecast reaches a strategist, because a confident wrong number is the most expensive artifact this organization can produce.
Standing duties include the miss review and the reproducibility standard: every forecast that missed gets a documented post-mortem, and every model gets a method note precise enough that a peer could rebuild it. This discipline is not bureaucracy — it is how a forecasting operation keeps its 89% historical accuracy record across 18 countries instead of slowly degrading into storytelling.
Scientists here also own the teaching surface of the cell: supervising intern models, reviewing their method notes, and feeding conversion decisions. The scientists who mentor well are promoted first, because modeling capacity at CryptoMize grows only as fast as its mentorship culture.
02
02Capability profile
What skills and tools does a Data Scientist need?
astro-island,astro-slot,astro-static-slot{display:contents}
Craft skills12 Tools & systems5 Offer standards4
Production-grade Python — pandas, scikit-learn, statsmodels pipelines others can runStatistical rigor — hypothesis testing, regression, Bayesian basics, and validation strategy that survives adversarial dataMachine learning in depth — gradient boosting, clustering, classification, and knowing which model a problem deservesTime-series forecasting — trend/seasonality decomposition, horizon honesty, backtesting disciplineExperimental design — A/B and quasi-experimental frameworks in messy field conditionsFeature engineering against dirty, manipulated, real-world corporaAdvanced SQL on multi-GB engagement warehousesModel communication — explaining uncertainty to a strategist who must act on itReproducibility engineering — seeds, versions, and audit trailsMentorship of interns and junior analystsDiscretion with NDA-grade client datasets (non-negotiable)Domain fluency across politics, reputation, security, and finance — the subject changes weekly
Python 3.11 (scikit-learn, statsmodels, XGBoost-class boosters)Prophet and classical time-series toolchainsPostgreSQL 15Git-based model repos with review disciplineJupyter production handoff standards
Depth of demonstrated skill in the specific role disciplineClassification and scope of the client engagement the role supportsUrgency and time-sensitivity of active project requirementsTrack record built across CryptoMize engagements
Also known as: data expert · machine learning specialist · big data analytics · analytics lead
03
03Seniority ladder
Data Scientist — seniority path at CryptoMize
Scientists advance by the quality of decisions their models enable under real conditions, not by tenure or publication count.
Data Scientist
Owns models and forecasting products for one to two engagements, from framing through validation to the client-facing number.
1/4
Senior Data Scientist
Designs the modeling architecture for new engagements, leads miss reviews, mentors interns, and signs off on methodology.
2/4
Principal Scientist / Forecasting Lead
Carries the accuracy record itself — standards, staffing, and accountability for every forecast CryptoMize puts its name on.
3/4
Intelligence Strategist
The crossover track: scientists who move into engagement strategy, translating models into the counsel clients act on.
4/4
04
04The engagement surface
CryptoMize work a Data Scientist touches
Every role plugs into live engagements across the five Penta-P domains — these are the services your work feeds.
Data Scientist · Job Opening
This seat plugs into 6 live CryptoMize services across the five Penta-P domains — the work below is where yours lands.
6 SERVICESPENTA-P
Big Data Mining
Predictive Intelligence
Strategic Intelligence
Sentiment Analysis
Political Surveys
OSINT
WHAT DOES DATA SCIENTIST COMPENSATION DEPEND ON?
Compensation is discussed during screening — never a fixed public figure, because it varies per person and per engagement. It depends on:
# OFFER CONSTRUCTION FACTOR
01 Depth of demonstrated skill in the specific role discipline
02 Classification and scope of the client engagement the role supports
03 Urgency and time-sensitivity of active project requirements
04 Track record built across CryptoMize engagements
