Artificial Intelligence Analyst

CryptoMize; New Delhi; On-site; INR 1,200,000 - 1,800,000; Full time

Job summary

CryptoMize in New Delhi is seeking an Artificial Intelligence Analyst to work on live intelligence problems, applying machine learning to multilingual sentiment streams, electoral histories, media-mention floods and coordinated-behavior traces. You will own models end-to-end: framing the question, engineering the features, validating honestly, and explaining results to strategists who will act on them.

The complete job description follows.

Responsibilities

• Build and run learning models on engagement corpora—sentiment trajectories, movement prediction, anomaly and coordination detection
• Frame engagement questions as learnable problems: what to predict, from what signal, validated against ground truth
• Engineer features from multilingual media, social and survey data to support decision tools

Skills

ML fundamentals

DL literacy

NLP multilingual

Feature engineering

Validation

Python for ML

Tools

Python

Jupyter

PostgreSQL

Experiment tracking

Job description

Artificial Intelligence Analyst jobs in Delhi at CryptoMize are open on a rolling, always-hiring basis. This is a full-time, permanent position with immediate joining at our New Delhi HQ, applying machine learning to the practice’s live intelligence problems. The corpora here are unlike textbook data: multilingual sentiment streams, electoral histories across 18 countries, media-mention floods and coordinated-behavior traces. Analysts build and run the models that turn those into foresight — feeding a forecasting record the practice stakes its name on. You will own models end-to-end: framing the question, engineering the features, validating honestly, and explaining results to strategists who will act on them. The complete job description — responsibilities, requirements, seniority path and selection process — follows.

What you will actually do:

• Build and run learning models on engagement corpora — sentiment trajectories, movement prediction, anomaly and coordination detection
• Frame engagement questions as learnable problems: what to predict, from what signal, validated against what ground truth
• Engineer features from messy multilingual media, social and survey data — the difference between a demo and a decision tool

Skills that matter:

Machine learning fundamentals, Deep learning literacy, NLP on noisy multilingual corpora (tokenization, embeddings, transfer learning), Feature engineering from media, social and behavioral data, Rigorous validation, Python for ML: pandas, scikit-learn, PyTorch/TensorFlow.

Tools you will use:

Python 3.11, Jupyter-based analysis repos with Git review discipline, PostgreSQL for feature and training data, Experiment tracking.

Location: New Delhi (HQ), E4, Vasant Vihar. Type: full-time permanent. Availability: immediate, rolling intake. Compensation: discussed at screening.