Functieomschrijving
Описание:
Booking.com develops travel-related chatbots, text generation and summarization applications, Q&A systems, free-text search, and travel-domain large language models with advanced agentic capabilities. Its solutions help redefine how travelers plan, book, and experience their journeys.
Задачи:
- Design, build, and maintain scalable infrastructure for training and fine‑tuning large language models on Booking.com's extensive data
- Optimize model serving and inference pipelines for latency, throughput, and cost efficiency at production scale
- Develop and maintain ML pipelines for data processing, model training, evaluation, and deployment
- Implement model optimization techniques such as quantization, distillation, pruning, and efficient attention mechanisms to meet production requirements
- Build monitoring, alerting, and observability systems for deployed ML models, ensuring reliability and performance in production
- Collaborate closely with ML scientists to translate research prototypes into production‑grade systems
- Contribute to reusable ML frameworks, tools, and libraries that accelerate the team's velocity
- Ensure code quality, scalability, and maintainability through testing, code reviews, and documentation
Требования:
- Strong software engineering skills with deep experience building and deploying ML systems at scale
- Hands‑on experience with LLM training infrastructure, including distributed training, GPU clusters, PyTorch, DeepSpeed, FSDP, or Megatron‑LM
- Experience with model serving and optimization using vLLM, TensorRT, ONNX, Triton Inference Server, or similar tools
- BSc with 4 years of work experience, or MSc with 3 years of work experience, in software engineering focused on machine learning systems
- Bachelor's, Master's degree, or equivalent experience in Computer Science, Software Engineering, or a related quantitative field
- Strong proficiency in Python
- Experience with cloud infrastructure, Kubernetes, Docker, AWS/GCP, and distributed computing frameworks such as Spark or Ray
- Familiarity with ML experiment tracking, CI/CD for ML, and data versioning tools
- Experience with large‑scale data processing pipelines using Kafka, Hadoop, Spark, Airflow, or similar tools
- Understanding of NLP and LLM concepts
- Ability to collaborate effectively with ML scientists and cross‑functional teams
- Excellent English communication skills, both written and verbal
- Proven ability to work in a fast‑paced, collaborative environment
- Nice to have: experience with Java, C++, or CUDA.
Условия:
- Annual paid time off and generous paid leave, including 22 weeks of paid parental leave, grandparent leave, bereavement leave, and care leave
- Flexible working arrangements, work‑from‑home furniture, ergonomic support, and up to 20 days per year working from abroad in the home country
- Sustainable HQ campus in Amsterdam with on‑site meals, coffee, snacks, multi-faith rooms, and breastfeeding rooms
- Commuting allowance and bike reimbursement scheme
- Discounts and wallet credits for Booking.com products, Booking.com Genius Level 3 upgrade, and friends and family discount vouchers
- Free access to online learning platforms, development and mentorship programs
- Global Employee Assistance Program and free Headspace membership
- Reasonable adjustments are available for individuals with disabilities during the interview process
- Equal opportunity workplace and affirmative action employer
- Pre‑employment screening may be conducted by a third party if the application is successful.