AstroTalk Data Scientist Recruitment 2026: Noida Job for 1–2 Years’ Experience
AstroTalk is hiring a Data Scientist for its Noida, Uttar Pradesh, location. The role is intended for candidates with 1–2 years of experience in data science or machine learning who have practical skills in predictive modelling, Python, SQL and working with customer or behavioural data.
The position focuses on solving business and product problems using machine learning. Responsibilities include building predictive models, exploring text data, engineering useful features, evaluating model performance and collaborating with product, business and engineering teams to turn data into actionable outcomes.
Candidates with experience using XGBoost, Random Forest, Logistic Regression and Natural Language Processing (NLP) should pay particular attention to this opening.
Job overview
| Detail | Information |
|---|---|
| Company | AstroTalk |
| Job title | Data Scientist |
| Location | Noida, Uttar Pradesh, India |
| Experience required | 1–2 years |
| Job category | Data Science / Machine Learning |
| Key skills | Python, SQL, Predictive Modelling, XGBoost, NLP |
| Posted on | 8 October 2026 |
| Employment type | Not explicitly stated in the supplied listing |
| Salary | Not specified |
| Application deadline | Not specified |
What will you work on?
This is a modelling-oriented role, with an emphasis on applying machine learning to customer, behavioural, transactional and product data. The goal is to identify patterns and develop solutions that help address real business problems.
Your work may cover several stages of the machine learning lifecycle:
Predictive modelling: Build and experiment with models for different business and customer use cases. The listing names XGBoost, Random Forest and Logistic Regression as examples of relevant algorithms.
Data exploration and feature engineering: Examine available data, understand its patterns and prepare meaningful features before training a model.
Model evaluation and optimisation: Assess how well a model performs, compare approaches and improve the solution where needed.
Text and NLP applications: Work with text data and explore relevant natural language processing use cases.
Cross-functional collaboration: Work with Product, Business and Engineering teams to turn data-driven ideas into practical solutions rather than leaving the results as isolated experiments.
The exact business problems and datasets are not specified in the job description, so candidates should not assume that the role is limited to a particular prediction task.
Skills that matter most
Python and SQL
These are core requirements. Candidates should be comfortable writing Python for data analysis and model development, and using SQL to retrieve and examine data. Be prepared to explain how you have used both tools in a real project.
Predictive modelling and XGBoost
The listing specifically asks for hands-on predictive modelling experience and names XGBoost. It also mentions Random Forest and Logistic Regression.
For preparation, review how to select an appropriate model, build a reliable validation approach, interpret evaluation metrics and recognise overfitting. You should be able to explain why you selected a particular algorithm instead of simply listing models on your resume.
Statistics and feature engineering
A good understanding of statistics, machine learning algorithms, feature engineering and model evaluation is expected. Candidates should be able to explain how they explored a dataset, handled relevant data-quality issues and decided which features could help solve the problem.
NLP and customer data
Practical experience with NLP and customer or behavioural data is explicitly mentioned. B2C, consumer internet or product-company experience is preferred, but the listing does not make that background mandatory.
Candidates from other environments should focus on demonstrating transferable experience with data-driven problem-solving and practical model development.
How to prepare a strong application
Instead of presenting a long list of libraries, choose one or two projects that demonstrate the modelling lifecycle from beginning to end.
For each project, be ready to explain:
- The problem: What were you trying to predict or understand?
- The data: What information did you use, and how did you prepare it?
- The features: Which variables mattered and why?
- The model: Why did you choose your approach?
- The evaluation: Which metrics did you use, and what did they reveal?
- The outcome: What improved, what did not work and what would you change next?
A project involving customer behaviour, transaction patterns or text classification could be particularly relevant if it genuinely reflects your experience. Do not claim that a project was deployed or produced business impact unless you can substantiate that claim.
Who should apply?
This opening is most relevant to candidates who:
- Have 1–2 years of data science or machine learning experience.
- Can work confidently with Python and SQL.
- Have built predictive models, particularly with XGBoost.
- Understand statistics, feature engineering and model evaluation.
- Have practical exposure to NLP and customer or behavioural data.
- Enjoy experimenting with different approaches to open-ended problems.
Experience in a B2C, consumer internet or product-based company is preferred. The supplied listing does not specify a mandatory degree, minimum CGPA, exact salary or detailed selection process.