Coherent Solutions ML Engineer role focused on classical machine learning initiatives including dataset preparation, model building, and deployment. Collaborate with senior engineers to develop, test, and deliver ML models and pipelines.
Responsibilities
Python, pandas, NumPy, scikit-learn, Jupyter
XGBoost, LightGBM
Git, Docker, pytest
Prepare and validate structured datasets for machine learning tasks, including cleaning, transformation, encoding, scaling, and missing-value handling;
Conduct exploratory data analysis to identify patterns, data quality issues, and opportunities for modeling;
Build, test, and compare ML models for tasks such as regression, classification, clustering, dimensionality reduction, and anomaly detection;
Develop and select features with guidance from senior engineers;
Run experiments, tune model parameters, and evaluate results using appropriate validation approaches and metrics;
Help maintain reproducible ML pipelines, notebooks, experiment tracking, and technical documentation;
Write and maintain automated tests for data and model-related code;
Support the delivery of models through REST APIs or batch-processing workflows;
Document model assumptions, results, limitations, and technical decisions;
Qualification
Strong foundation in statistics
Required
1+ year of hands-on Python development experience, including experience with pandas, NumPy, scikit-learn, and Jupyter
Strong foundation in statistics, probability, and linear algebra, plus exploratory data analysis skills
Understanding of supervised and unsupervised learning, and practical experience with regression, classification, clustering, dimensionality reduction, and anomaly detection
Experience preparing structured datasets, including cleaning, transformation, encoding, scaling, and missing-value handling, plus practical feature engineering and feature-selection skills
Understanding of train/validation/test splits, cross-validation, data leakage, overfitting, and regularization
Ability to select appropriate evaluation metrics, compare models, and perform error analysis
Familiarity with hyperparameter tuning and reproducible ML pipelines
SQL skills for data extraction and analysis
Ability to expose models through REST APIs or batch-processing workflows