Advanced Machine Learning Engineer

Role Summary

We are seeking an Advanced Machine Learning Engineer to design, build, train, deploy, and operate scalable machine learning systems. This hands-on ML engineering role focuses on implementing models and production systems—not solely conducting data analysis or research. It requires strong Python and deep-learning expertise, experience with frameworks such as PyTorch or TensorFlow, and the ability to own the end-to-end model lifecycle, from behavioral/event data preparation and feature engineering through model development, large-scale training, deployment, inference optimization, and monitoring. The engineer will work with Snowflake, Databricks, Spark, and GPU-based training infrastructure to deliver reliable, efficient, and maintainable ML capabilities.

Must have - Buulding and operated production ML systems using Python + PyTorch /TensorFlow + Snowflake/Databricks + MLOps.

Key Responsibilities

Machine Learning Engineering

Implement and optimize machine learning algorithms and models for business use cases, including classification, regression, clustering, and recommendation systems as applicable.

Translate model requirements into robust, reusable, and testable Python components.

Build model training, validation, scoring, and inference workflows that can be integrated into production applications.

Evaluate model performance against agreed technical and business metrics and implement improvements to accuracy, latency, robustness, and resource efficiency.

Apply appropriate validation, error analysis, and model testing practices to ensure reliable behavior.

Train models efficiently on GPU infrastructure using techniques such as mixed-precision training, gradient accumulation, checkpointing, memory optimization, and distributed training strategies where appropriate.

Use PyTorch, TensorFlow,Pyspark or equivalent frameworks to implement experiments and production-ready training code; apply rigorous evaluation, error analysis, and reproducibility practices.

Build model-training workflows covering event/token sequence preparation, tokenization, temporal and positional representations, self-supervised pretraining, and downstream fine-tuning when appropriate to the use case.

Design, implement, train, and optimize deep-learning models; develop custom architectures where required, including transformer- and attention-based models for sequential or behavioral data.

Deep Learning & Model Architecture

Data Platforms & ML Pipelines

Use Snowflake and SQL to access, query, and prepare enterprise data required by machine learning applications.

Use Databricks for scalable data processing, feature preparation, model development, and ML pipeline execution.

Develop and maintain reliable pipelines for data preparation, feature generation, training, batch scoring, and inference.

Use PySpark and distributed processing techniques where appropriate for large-scale workloads.

Work with data engineering teams to integrate upstream data sources and ensure dependable data contracts, quality checks, and pipeline operation.

Productionization & MLOps

Productionize ML solutions and integrate them with services, APIs, data products, or downstream business workflows.

Implement model and code versioning, experiment tracking, reproducible builds, automated testing, and release practices.

Monitor model outputs and pipeline health; troubleshoot failures, data changes, drift, and performance degradation.

Improve operational reliability, scalability, observability, and maintainability of ML systems.

Collaborate with platform, cloud, and DevOps teams on deployment automation and production support.

Build observable production ML systems with appropriate logging, alerting, performance tracking, and operational safeguards.

Optimize inference latency, throughput, memory use, and serving cost through model architecture choices, batching, and serving-side optimizations.

Software Engineering & Technical Ownership

Develop clean, modular, documented, and maintainable code following engineering standards and design principles.

Contribute to technical design, code reviews, testing strategies, and architectural decisions for ML systems.

Own assigned ML engineering deliverables from technical design through implementation, release, and ongoing support.

Identify performance bottlenecks and improve runtime, resource utilization, and processing efficiency.

Collaborate with cross-functional engineering and product teams to deliver dependable ML capabilities.

Required Skills & Qualifications

Strong proficiency in Python and software engineering practices, including object-oriented design, modular development, testing, debugging, and code quality.

Strong ML foundation and mathematical understanding

Strong understanding of the end-to-end ML lifecycle, including model architecture, training, evaluation, optimization, inference, production integration, and monitoring.

Strong SQL skills and experience working with structured data and data pipelines.

Hands-on experience with Snowflake for querying and preparing data for ML workflows.

Hands-on experience with Databricks for data processing and/or ML pipeline development.

Strong hands-on experience with PyTorch, TensorFlow, or a similar deep-learning framework; experience with scikit-learn is also useful where appropriate.

Understanding of production ML concepts, including reproducibility, model versioning, deployment, monitoring, and troubleshooting.

Ability to design reliable, scalable, and maintainable software components and pipelines.

Familiarity with GPU training concepts and distributed strategies such as Distributed Data Parallel (DDP), Fully Sharded Data Parallel (FSDP), or equivalents is an advantage for large-model workloads.

Understanding of model optimization, inference efficiency, and integration with production services or applications.

Experience developing and deploying ML solutions at scale, including end-to-end model lifecycle