ML Engineer (Onsite, Lahore)

Requirements:

• Strong hands-on experience with classification models and machine learning libraries such as XGBoost, scikit-learn, and other Python-based ML tools.
• Proficient in Python programming with solid experience in data handling libraries like Pandas.
• Experienced in working with text data, including preprocessing, tokenization, feature extraction (bag-of-words, TF-IDF, embeddings), and integrating text features into traditional ML models.
• Familiar with NLP libraries such as spaCy, NLTK, HuggingFace Transformers, and dimensionality reduction techniques like PCA, SVD, and UMAP for text.
• Hands-on experience with neural networks and a strong understanding of ML fundamentals, including feature engineering, model evaluation, hyperparameter tuning, bias-variance tradeoff, and interpretability.
• Experienced in MLOps practices such as model versioning, deployment, monitoring, lifecycle management, and retraining pipelines using tools like MLflow or Weights & Biases.
• Skilled in AWS services relevant to ML workflows especially Lambda, S3, and SageMaker.
• Proficient in reading, understanding, and improving production-grade Python codebases with familiarity in version control (Git), code reviews, and agile development practices.
• Comfortable with structured data systems and handling large datasets using SQL, BigQuery, or similar tools.
• Strong communication skills with the ability to present technical concepts to both technical and non-technical stakeholders.
• A proactive problem-solver and critical thinker who seeks continuous improvement, identifies gaps, and leads new initiatives.
• A collaborative team player who combines strong analytical thinking with a hands-on approach to both optimizing existing models and building innovative solutions.
  

Responsibilities:

• Deeply understand the current ML models architecture, feature engineering pipeline, and training methodology with a strong focus on text-based data inputs.
• Perform critical evaluations of existing feature engineering and model decisions.
• Analyze model performance, identify opportunities for optimization, and address technical debt.
• Propose and implement improvements to feature processing pipelines, including better handling of text features (e.g., TF-IDF, embeddings, and dimensionality reduction).
• Lead the design and development of new machine learning models or approaches as the product evolves.
• Maintain and improve model training, validation, and deployment pipelines for production.
• Collaborate closely with data scientists, software engineers, and product teams to align modeling strategies with business objectives.
• Ensure robust documentation of model internals, evaluation findings, and improvement strategies.