Excelacom Technologies is looking for an experienced Machine Learning Data Scientist with robust expertise in Anomaly Detection to lead the design, development, and deployment of advanced machine learning solutions. The ideal candidate will have hands-on experience in building production-grade ML models, managing teams, and working directly with clients to solve complex business problems.

Candidates with proven experience in Anomaly Detection are highly preferred.

Company: Excelacom Technologies Pvt. Ltd.
Location: Siruseri, Chennai
Work Mode: Work from Office
Experience: 5 20 Years
Notice Period: Immediate Joiners / Maximum 15 Days Preferred

Key Responsibilities

Data Exploration & Engineering
Collect, clean, preprocess, and transform structured and unstructured datasets.
Perform exploratory data analysis (EDA) to identify patterns, trends, and anomalies.
Engineer features for supervised, unsupervised, and time-series models.
Work with enterprise datasets including logs, transactional data, sensor/IoT data, and CRM data.
Implement data validation, quality checks,



and data augmentation pipelines.

Model Development & Evaluation
Design and develop machine learning models for classification, regression, clustering, and anomaly detection.
Apply anomaly detection techniques such as:
Isolation Forest
Autoencoders
One-Class SVM
Time-Series Anomaly Detection
Evaluate models using metrics such as Precision, Recall, F1 Score, AUC-ROC, RMSE, etc.
Manage model drift, retraining strategies, and continuous model improvement.
Work on RAG pipelines and vector search optimization for GenAI-powered analytics.

Machine Learning & Deep Learning Engineering
Build ML/DL pipelines using Python, PyTorch, TensorFlow, and Scikit-learn.
Develop NLP and time-series forecasting solutions.
Optimize models for performance, latency, scalability, and cost efficiency.
Integrate ML solutions into AI automation frameworks such as LangChain and CrewAI.

Team Lead

📌 Urgent Openings For Ml Data Scientist Chennai
🏢 ExcelaCom Technologies
📍 Chennai