Role Overview

At Haber, we are changing how industries run by combining automation with intelligent decision making. As a

Senior Data Scientist, you will build and own the machine learning systems behind our products - Elixa, Kaiznn,

Mount Fuji and our upcoming Fibre Morphology solution.

You will work with time-series sensor data coming from live plants, and build models for quality prediction,

process optimisation, anomaly detection, forecasting and root cause analysis. You will also lead a small team of 2

to 4 data scientists and set the technical direction for how we build and deploy models.

This is a hands-on leadership role. Our models do not stay in notebooks. They run in operating plants, where a wrong prediction costs real money, water and energy. We are looking for someone who has built a model, put it in production, seen it degrade, and fixed the system behind it.

Key Responsibilities

● Build AI and ML solutions for quality prediction, process optimisation, anomaly detection, forecasting, root cause analysis and decision support

● Work with time-series sensor data from live plants, and handle the real problems that come with it - sensor drift, missing data, changing conditions and limited labels

● Own the full lifecycle: data collection, cleaning, feature engineering, model building, validation, explainability,

deployment and monitoring

● Take models into production and keep them healthy - pipelines, versioning, monitoring, drift detection,

retraining and rollback

● Own the technical direction for the data science team and decide what gets built, what gets reused and what gets retired

● Lead and mentor 2 to 4 data scientists through code reviews, design reviews and day-to-day guidance

● Contribute to AI system design - data access layers, model services, APIs and cloud deployment on AWS





● Build optimisation and recommendation layers that act on model output in real time, such as setpoint recommendation for plant operations

● Work on Generative AI and Agentic AI solutions using LLMs, RAG, tool calling and knowledge graphs, as these become part of our product

● Work closely with domain experts, product, software, cloud and engineering teams to turn industrial problems into working solutions

● Stay current with research in machine learning, signal processing and industrial optimisation

Must Have

● 8 to 10 years of experience in Data Science and Machine Learning, with real production experience

● Bachelors or Masters in Engineering, Computer Science, Data Science, Mathematics, Statistics or a related field

● Strong Python and SQL, with Pandas, NumPy and Scikit-learn

● Strong foundation in statistics, probability, mathematics and optimisation

● Strong experience in time-series analysis and forecasting - this is the core of our work

● Machine learning: regression, classification, clustering, anomaly detection and ensemble models

● Proven experience deploying and maintaining models in production, including monitoring, drift detection and retraining

● MLOps and production tools: Docker, Git, CI/CD, MLflow, FastAPI or similar

● AWS for deployment and monitoring - S3, Lambda, SageMaker, CloudWatch

● Working knowledge of deep learning using PyTorch or TensorFlow, especially for time-series problems

● Model explainability using SHAP, feature importance or similar methods





Haber – Senior Data Scientist Page 1● Experience mentoring or leading other data scientists

● Ability to work independently and take a project from idea to production

● Good communication skills and comfort working with people who are not data scientists

Good to Have

● Experience in manufacturing, chemicals, pulp and paper, water, energy, pharma, metals or similar industrial and process environments

● Generative AI: LLMs, prompt engineering, RAG, embeddings and vector databases

● Agentic AI: multi-agent systems, tool and function calling, and Model Context Protocol (MCP)

● LangGraph, LangChain or similar frameworks; Knowledge Graphs or Neo4j

● AWS Bedrock and RDS

● Computer vision: image preprocessing, object detection, classification, segmentation, OpenCV, YOLO, CNNs or Vision Transformers

● Optimisation, control theory, reinforcement learning or operations research

● Signal processing fundamentals

● Causal analysis, industrial digital twins, or AI governance and ISO 42001 exposure

● Experience building system design, APIs and microservices for production AI systems

What We Offer

● Real-World Impact: Your models will run in operating plants, not just in experimental labs

● Ownership and Autonomy: Freedom to explore, decide and build in a high-trust environment

● Leadership: A small team from day one, with room to grow into wider data science leadership

● Diverse Tech Stack: Time-series, sensor data, vision, optimisation and generative AI in one place

● Quick-Track Learning: A startup where learning never stops and innovation is constant

● Collaborative Culture: A team that supports initiative, experimentation and personal growth

● Product-Minded Thinking: Your work will be closely tied to product and business decisions

📌 Senior Data Scientist (Pune)
🏢 Haber
📍 Pune