# Data Scientist Internship at CryptoMize Apply now for the Data Scientist Internship in Delhi at CryptoMize — a 6-month certified program on rolling, always-hiring intake. This is not a notebooks-only internship: CryptoMize runs one of the few applied forecasting shops where models genuinely face elections, reputation crises, and threat landscapes across 18 countries, and Data Scientist interns work inside that pipeline from week one. You will build clustering, classification, and time-series models on real engagement corpora, backtest them against outcomes we have already recorded at 89% historical accuracy, and learn what the textbook quietly omits — that data in the wild arrives dirty, adversarial, and mid-crisis. Below is the full internship description: required skills, responsibilities, eligibility, and the selection process. ## Facts - Location: New Delhi (HQ), CDR Building, Chhatarpur - Type: 6-month internship, certification provided - Availability: immediate, rolling intake (always hiring) - Compensation: discussed at screening ## What you will actually do - Build and backtest forecasting models — voter-preference, sentiment-trajectory, and risk-signal — against CryptoMize’s recorded engagement outcomes, and document every miss honestly - Run the full modeling loop on real corpora: cleaning, feature engineering, model selection, validation, and the written method note a strategist can audit - Experiment with clustering and dimensionality reduction to segment audiences for perception and political engagements (unsupervised work the platforms actually consume) - Support A/B and quasi-experimental frameworks on reputation interventions, learning to measure movement without fooling yourself - Work alongside the data engineering side — the pipelines you model on are the pipelines you learn to respect - Present model results in the Friday review, where a strategist will ask "is that signal or noise?" and teach you how to know - Sit inside the applied end of an 89%-accuracy forecasting operation across 18 countries — the internship where machine learning meets consequence - Learn the production hygiene of real modeling work — versioned datasets, pinned environments, and handoff notes that let someone else rerun your experiment months later ## Skills - Python for data science — pandas, NumPy, scikit-learn in genuine working depth - Statistical foundations — distributions, hypothesis testing, regression, overfitting discipline - Machine learning fundamentals — clustering, classification, gradient boosting, validation strategy - Time-series basics — trend, seasonality, and the humility of forecasting at horizon - Feature engineering on messy, real-world corpora (not Kaggle-clean CSVs) - SQL for extraction and joins against multi-table engagement data - Experimental design — A/B structure, control groups, and quasi-experimental reasoning - Data visualization for model communication — explaining a model to a non-modeler - Reproducibility discipline — notebooks others can rerun, seeds and versions pinned - Written method documentation — the audit trail behind every number - Confidentiality discipline with NDA-grade client datasets - Intellectual honesty — shipping the model that is right, not the one that is impressive ## Tools - Python 3.11 (pandas, scikit-learn, statsmodels) - Prophet for time-series forecasting - Jupyter Notebooks with versioned outputs - PostgreSQL for extraction - Git-based reproducible analysis repos ## Questions and answers ### What does a Data Scientist intern do at CryptoMize? A Data Scientist intern builds, validates, and backtests forecasting and segmentation models on real client engagement data — electoral, sentiment, media, and threat corpora — and documents the method behind every result. The work feeds live engagements across 18 countries, not classroom datasets. ### What is the eligibility for the Data Scientist Internship? Final-year students and recent graduates in statistics, mathematics, computer science, or equivalent self-taught evidence. We evaluate working Python and statistics depth, not paperwork — no diploma is mandatory at CryptoMize. ### Which skills are required for a Data Scientist internship? Working Python (pandas, scikit-learn), solid statistics, machine-learning fundamentals including clustering and classification, SQL for extraction, and the discipline to document what a model actually does. Time-series exposure helps; we teach the rest in-house. ### Is the Data Scientist Internship paid? Compensation is discussed during screening and depends on demonstrated skill, engagement classification, urgency, and track record. We do not publish figures — every intern is evaluated individually. ### How long is the internship, and is a certificate provided? Six months at our New Delhi HQ, with a certificate on completion — and a portfolio of real model work against recorded outcomes. Strong interns receive full-time Data Scientist offers. ### What is the selection process for the Data Scientist Internship? A short application, one modeling-and-SQL screening task on a realistic dataset, and a final interview with the analytics lead where you defend your modeling choices. The cycle typically completes within two weeks. ### Is this a data science internship in Delhi or remote? The internship is based at our New Delhi HQ in Vasant Vihar, working directly with the forecasting team whose models face live engagements. ### Is this a data science internship or a data analytics one? Data science: the emphasis is modeling — forecasting, clustering, classification, and experimental measurement — rather than reporting. You will sit next to the Data Analyst interns and learn their craft too, but your deliverables are models and method notes, not dashboards. Apply: https://cryptomize.com/careers/internship/apply/