Machine Learning Engineer - AI

Accountabilities:

• Fine-tune and train Small Language Models (SLMs) using Hugging Face, TRL, and parameter-efficient adaptation techniques such as LoRA, QLoRA, and PEFT.
• Experiment with model architectures, training strategies, and datasets to improve model quality and task-specific performance.
• Optimize models for efficient inference using techniques including quantization, pruning, knowledge distillation, and other model compression approaches.
• Prepare and deploy lightweight AI models to edge devices, mobile environments, local servers, and other resource-constrained platforms.
• Design and implement end-to-end MLOps pipelines covering data ingestion, preprocessing, experimentation, model training, validation, packaging, deployment, and monitoring.
• Build reliable and repeatable workflows that support efficient model development and production deployment.
• Monitor deployed models for accuracy, latency, resource consumption, and CPU/GPU utilization.
• Develop and maintain model benchmarking frameworks and custom evaluation suites to measure model quality and performance.
• Analyze production performance and identify opportunities to improve model efficiency, reliability, and scalability.
• Work with cross-functional engineering teams to integrate machine learning models into real-world products and environments.
• Contribute to ML engineering best practices around experimentation, versioning, deployment, observability, and continuous improvement.
• Explore emerging techniques and tooling for efficient AI inference, edge deployment, and production machine learning.

Requirements:

• Hands-on experience developing, training, and fine-tuning Small Language Models or other transformer-based models.
• Strong practical knowledge of Hugging Face and modern model adaptation techniques, including LoRA, QLoRA, and PEFT.
• Experience optimizing machine learning models for efficient inference through quantization, pruning, knowledge distillation, or similar techniques.
• Experience deploying machine learning models to edge devices, mobile platforms, local servers, or other environments with constrained compute and strict latency requirements.
• Strong understanding of end-to-end MLOps practices, from data ingestion and model experimentation through deployment and production monitoring.
• Experience monitoring model accuracy, inference latency, and CPU/GPU or other hardware utilization in production.
• Ability to develop meaningful model evaluation and benchmarking frameworks and use data-driven results to improve model performance.
• Strong software engineering, debugging, analytical, and problem-solving skills.
• Ability to work effectively in a collaborative, fast-moving environment and communicate technical concepts clearly.
• Strong ownership mindset and willingness to continuously learn new machine learning technologies and deployment techniques.
• Experience with ONNX export and cross-platform inference is a plus.
• Experience deploying AI solutions to edge or mobile environments is preferred.
• Familiarity with MLOps tooling for experiment tracking, model registries, and ML-focused CI/CD pipelines is advantageous.

Benefits:

• Opportunity to work on modern AI and machine learning technologies with a strong focus on Small Language Models.
• Hands-on exposure to model fine-tuning, optimization, compression, and efficient inference.
• Opportunity to build production-grade MLOps pipelines spanning the full machine learning lifecycle.
• Experience working with edge, mobile, and local AI deployment environments.
• Exposure to technologies including Hugging Face, TRL, LoRA, QLoRA, PEFT, and modern MLOps tooling.
• Opportunity to contribute to scalable AI solutions designed for real-world production environments.
• Collaborative environment with opportunities to work alongside experienced engineering and technology professionals.
• Strong focus on continuous learning, experimentation, and adoption of emerging AI technologies.
• Inclusive workplace culture that values diverse perspectives, collaboration, and individual contributions.