Data Scientist Intern

Revrag Ai; Bengaluru; On-site; INR 167,000 - 279,000; Full time

Job summary

RevRag.ai is seeking a Data Scientist Intern to work on ML and data science powering our Voice and In-App AI Agents. You will analyze conversation data and usage patterns, build and tune models, and help improve agent quality in production.

This role focuses on applied data science using real BFSI data, turning messy datasets into actionable models, experiments, and insights that ship.

Qualifications

• Pursuing or recently completed degree in CS/DS/Statistics/Math.
• Strong foundation in ML and statistics with practical project work.
• Hands-on Python experience and model development.

Responsibilities

• Explore and analyze large conversation, call, and usage data to surface insights.
• Build, train, and evaluate ML models for classification and scoring.
• Fine-tune LLMs and related models for BFSI use cases.
• Design evaluation pipelines and quantify model performance.
• Collaborate with engineers to deploy models into production.

Skills

ML fundamentals

Python

PyTorch / TensorFlow

Data analysis

Education

Bachelor's or Master's in CS/DS/Statistics/Math

Tools

Pandas

NumPy

SQL

Scikit-learn

Job description

Job Summary

As a Data Scientist Intern, you will work on the machine learning and data science that powers RevRag AIs Voice and In-App AI Agents, from analysing conversation, call, and usage data to building, fine-tuning, and evaluating models that improve agent quality in production.

This role is focused on practical, applied data science, not academic research. Youll turn messy, real-world BFSI data into models, experiments, and insights that ship.

Key Responsibilities

• Explore and analyse large conversation, call, and usage datasets to surface patterns, insights, and improvement opportunities.
• Build, train, and evaluate ML models for classification, scoring, and quality tasks across the agent platform.
• Fine-tune and adapt LLMs and smaller models for specific BFSI use cases.
• Design evaluation pipelines and metrics to measure model and agent performance (accuracy, reliability, business impact).
• Read relevant AI and ML research and implement useful ideas from papers.
• Run structured experiments, track results, and document findings clearly.
• Work with engineers to integrate models into production workflows.
• Build data pipelines and tooling for feature engineering, labelling, and analysis.

Technical Skills

• Strong understanding of Machine Learning and Artificial Intelligence fundamentals, including supervised and unsupervised learning, deep learning basics, and common model architectures.
• Good grasp of the mathematical foundations behind machine learning models, including linear algebra, probability, statistics, calculus, and optimization techniques.
• Comfortable reading and understanding AI and ML research papers and implementing ideas from them when required.
• Hands-on experience with model fine-tuning through academic projects, internships, research work, or personal projects.
• Familiarity with Python and popular ML frameworks such as PyTorch, TensorFlow, or Scikit-learn.
• Analytical mindset to evaluate model outputs and data quality.
• Currently pursuing or recently completed a degree in Computer Science, Data Science, Statistics, Mathematics, or a related field.

Nice to Have

• Published or co-authored AI and ML research papers, or active contributions to open-source AI projects.
• Experience participating in AI and ML hackathons, research initiatives, or Kaggle competitions.
• Exposure to LLMs, embeddings, and vector databases.
• Experience with data tools (Pandas, NumPy, SQL) and visualization.
• Exposure to cloud platforms (AWS, GCP).

What Youll Learn

• How to apply data science and ML in real production AI systems.
• Fine-tuning and evaluating models for real BFSI use cases.
• Turning messy real-world data into models and insights.
• Building evaluation pipelines and measuring what actually matters.
• How models behave under real user conditions.

Success in This Role

• Ship measurable improvements to model or agent performance.
• Deliver clean, well-documented experiments and analyses.
• Turn data into decisions that influence the product.
• Demonstrate structured experimentation and strong learning velocity.

Disclaimer: This job posting has been aggregated from external source. Role details, content, and availability are subject to change. Applicants are advised to confirm the latest information directly on the company website before applying.