Machine Learning Engineer

Remote Position

Must have languages: Python, SQL, (Scala or Java).

Must have skills: PyTorch, Huggingface transformers, Deep Learning, Spark/Hadoop, MLOps

Additional must haves: Strong communication skills. Must be willing to contribute to team discussions and solutions. Also, must be available for stand ups, core meetings, and slack availability.

Team Overview

Advertising is a rapidly expanding and strategically vital sector that is shaping our organisation s future. The digital advertising industry is evolving quickly, with a notable shift as ecommerce advertisers recognize the significant value offered by platforms. This creates a massive opportunity for us as advertisers reallocate spending away from our competitors duopoly. Furthermore, advertising strengthens our core ecommerce business by providing sellers with a crucial tool to accelerate inventory movement and enhancing the buyer experience through the surfacing of high-quality items.

Team Role

This applied research team is dedicated to developing intelligent, end-to-end solutions that empower our advertiser sellers to achieve their advertising goals. This involves providing strategic guidance, such as recommending optimal ad placements, selecting the right inventory, keywords, bids, and budgets for their campaigns, and facilitating continuous optimization. This rapidly evolving space presents significant business potential and is a critical area of need. To address this, we build scalable data and Machine Learning/Deep Learning services, which entails working with massive datasets, applying diverse data science and ML/DL techniques, and adhering to the best engineering practices.

Responsibilities

• Work with Applied Researchers, Engineers, Analytics and multi-functional teams to produce end-to-end production-ready solutions.
• Design, implement, and maintain data and ML/DL services, encompassing efficient data pipeline, ML model training, inference & deployment processes, robust tracking and monitoring system, as well as other MLOps work.
• Optimize software performance to achieve the required throughput and / or latency.
• Analyze data, interpret experiments, and uncover trends, insights, and opportunities. Translate complex data into actionable recommendations for technical and non-technical audiences.
• Monitor, triage and drive resolution processes for production system incidents, ensuring system reliability.

Requirements

• BS or MS in Computer Science or equivalent experience
• Strong programming skills in SQL, Python, Scala or Java
• Solid understanding of machine learning fundamentals and applications.
• Proficiency with key machine learning and deep learning libraries, e.g. PyTorch/Tensorflow, Transformers, scikit-learn, vLLM, and Ray.
• Strong debugging, analytical, and communication skills to collaborate effectively with engineers and researchers.
• Good understanding of ML/DL infrastructure, including areas like autoscaling, job scheduling, and workload orchestration across heterogeneous compute (CPU/GPU/accelerators).
• Experience with big data technologies (i.e., Spark, Hadoop) and database technologies (i.e., SQL, NoSQL)
• Experience in solving problems using data science, building practical solutions, and deploying models into production environments.
• Experience with observability stacks (e.g. Prometheus, Grafana) and data visualization tools (e.g., Tableau, Kibana) is a plus.
• Experience with developing large language model (LLM) applications is a plus.