New opportunity

Senior Data Science Engineer

Sabre · Bengaluru, India

About this role

The Senior Data Science Engineer will join the Data Intelligence & Research team at Sabre, serving as the intelligent core of the company's travel marketplace. This role involves planning, designing, and deploying complex software solutions that integrate advanced mathematical algorithms and machine learning models into core platforms such as Airline Pricing, Network Planning, and Agency Automation.

Responsibilities

Architect, code, and deploy complex software solutions and data intelligence pipelines integrating mathematical algorithms and machine learning models.
Drive research and innovation by prototyping and scaling next-gen ideas, including AI-driven automation and real-time itinerary optimization.
Design end-to-end AI systems, focusing on LLM integration, customer intent recognition, and conversational commerce.
Establish strategic AI lifecycle policies and MLOps standards to ensure high-fidelity model performance.
Refactor research-grade prototypes into production-quality, low-latency systems using Python, Golang, or C++.
Collaborate with cross-functional leadership to align technical roadmaps with business goals like NDC adoption and operational efficiency.
Mentor junior data scientists and engineers while fostering a culture of rigorous experimentation.

Required qualifications

Minimum 4 years of related experience in Data Science, AI Engineering, or Operations Research.
Advanced degree (Masters/PhD preferred) in Mathematics, Machine Learning, Statistics, Computer Science, or Physics.
Proven experience in Agile software development and moving complex models from research to global production.
Expertise in advanced ML/AI solutioning, including Deep Learning, Reinforcement Learning, and NLP.
High-performance programming skills in Python, Golang, or C++.
Deep knowledge of the modern AI tech stack, including GCP (Vertex AI, BigQuery), MLOps, Docker, and Kubernetes.
Proficiency in Generative AI frameworks and RAG-based systems.

Preferred qualifications

Experience in the travel industry, specifically within GDS or OTA spaces.
Domain knowledge in media operations, including AdTech ecosystems, real-time bidding, or media mix modeling.

For full details about this role, click “Apply on company site” to view the employer’s original posting.

Skills for this role

PythonGolangC++Machine LearningDeep LearningReinforcement LearningNatural Language Processing (NLP)Generative AIRAGMLOpsGCPVertex AIBigQueryDockerKubernetesMathematical Optimization

Checking your profile…

Preparation suggestions, not the employer’s actual interview questions. Always check the original posting for current requirements.

Understand the role

What matters most

Productionizing Research-Grade AI Models
The core mandate is to refactor complex mathematical and machine learning prototypes into low-latency, production-quality systems. You must demonstrate the ability to bridge the gap between experimental research and high-availability, mission-critical travel platforms using Python, Golang, or C++.
Architecting Scalable MLOps and AI Lifecycle Policies
You are responsible for establishing strategic AI lifecycle policies and MLOps standards. This requires deep expertise in GCP, Vertex AI, Docker, and Kubernetes to ensure model performance remains high-fidelity despite the volatile data landscapes inherent in the travel industry.
Integrating Generative AI and LLMs into Travel Workflows
The role focuses on designing end-to-end systems that leverage LLMs, RAG-based architectures, and customer intent recognition. You must align these technical implementations with business goals like NDC adoption and conversational commerce.
Technical Leadership and Mentorship
Beyond individual contribution, you must foster a culture of rigorous experimentation and mentor junior team members. You will influence cross-functional stakeholders to adopt disruptive technologies like Agentic AI within the existing travel infrastructure.

Likely challenges to think through

Balancing Latency with Model Complexity
Travel systems require real-time responsiveness, yet you are tasked with integrating advanced Deep Learning and Reinforcement Learning models. Prepare to discuss how you optimize computationally intensive algorithms for low-latency environments without sacrificing predictive accuracy.
Navigating Legacy System Integration
A significant challenge involves migrating legacy automation logic into modern, cloud-native GCP architectures. Be prepared to explain your strategy for maintaining system stability during the transition and how you handle data consistency across hybrid environments.
Managing High-Volume, Volatile Travel Data
The travel industry is characterized by high-volume, real-time data streams. You will face the difficulty of ensuring model robustness against sudden market shifts or data anomalies; prepare to discuss your approach to data monitoring and automated retraining pipelines.

Practice your answers

Technical practice

Refactoring Research Prototypes for Production

What to coverExplain your process for transitioning a research-grade model into a low-latency production system. Focus on how you handle performance bottlenecks, code optimization in Python or Golang, and ensuring the system meets high-availability requirements.

Designing RAG Systems for Workflow Automation

What to coverDescribe the architecture of a RAG-based system you have built. Discuss how you manage document retrieval, context window limitations, and how you ensure the accuracy of the generated output for domain-specific tasks.

MLOps and Model Lifecycle Management

What to coverHow do you establish MLOps standards to maintain high-fidelity model performance in a volatile data environment? Suggestion: Focus on your experience with GCP tools like Vertex AI, automated retraining pipelines, and monitoring for data drift.

Mathematical Optimization in Travel Systems

What to coverDiscuss a scenario where you integrated classical mathematical optimization with machine learning. Explain the trade-offs you considered when balancing algorithmic complexity with the need for real-time response in systems like airline pricing or network planning.

Scaling AI Systems on GCP

What to coverDetail your experience deploying containerized AI solutions using Docker and Kubernetes on GCP. Address how you manage resource allocation for computationally intensive tasks and ensure scalability during peak travel demand periods.

Reinforcement Learning for Real-Time Optimization

What to coverProvide an example of applying Reinforcement Learning to a real-world problem. Explain how you defined the state space, reward function, and how you addressed the challenges of training stability in a production-like environment.

NLP and Intent Recognition

What to coverHow do you approach customer intent recognition in a conversational commerce context? Discuss the techniques you use to handle ambiguity and how you evaluate the performance of your NLP models.

High-Volume Data Processing

What to coverGiven the high-volume nature of GDS and OTA data, how do you design data pipelines to ensure low-latency access? Suggestion: Discuss your experience with BigQuery or similar distributed data warehouses and strategies for efficient data ingestion.

Behavioural practice

Cross-Functional Collaboration and Influence

What to coverDescribe a time you had to align a technical roadmap with business goals, such as NDC adoption or operational efficiency. Explain how you communicated complex AI concepts to non-technical stakeholders to secure their buy-in.

Mentoring and Fostering Experimentation

What to coverProvide an example of how you have mentored junior team members. Focus on how you encouraged a culture of rigorous experimentation while ensuring that project deadlines and quality standards were met.

Driving Innovation with Minimal Direction

What to coverTell me about a complex, cross-functional project you led from conception to completion. How did you identify the opportunity, manage dependencies, and overcome obstacles when working with minimal supervision?

Handling Technical Disagreement

What to coverDescribe a situation where you disagreed with a technical approach proposed by a peer or leadership. How did you advocate for your position, and what was the final outcome for the project?

Sharpen your knowledge

Topics to prepare

MLOps and AI Lifecycle Management
The role requires establishing strategic AI lifecycle policies and MLOps standards. Exercise: Review the MLOps maturity model and design a strategy for monitoring model drift and performance in a high-latency production environment. This is relevant to ensuring high-fidelity model performance in Sabre's travel marketplace.
Agentic AI and RAG Architectures
The position emphasizes Generative AI frameworks and RAG-based systems for workflow automation. Exercise: Architect a RAG pipeline that integrates real-time GDS data to answer complex travel queries. This is directly relevant to the company's focus on the agentic revolution in travel.
Mathematical Optimization in Travel
The role involves integrating mathematical algorithms into airline pricing and network planning. Exercise: Practice formulating a constrained optimization problem for airline itinerary scheduling. This is relevant to the core mission of optimizing travel distribution and operational efficiency.

Skills to revise

High-Performance Programming (Python, Golang, or C++)
The role requires refactoring research-grade prototypes into production-quality, low-latency systems. Exercise: Optimize a computationally intensive Python script by profiling it and identifying bottlenecks, or rewrite a core function in Golang or C++. This is essential for meeting the latency requirements of Sabre's platforms.
GCP Cloud Infrastructure (Vertex AI, BigQuery)
The job requires deep knowledge of the modern AI tech stack on GCP. Exercise: Design a data pipeline in BigQuery that feeds into a Vertex AI model for real-time inference. This is relevant to the requirement of migrating legacy logic into modern, cloud-native architectures.
Containerization and Orchestration (Docker, Kubernetes)
The role involves deploying complex software solutions. Exercise: Create a Dockerfile for a machine learning model and define a Kubernetes deployment manifest to manage its scaling. This is necessary for maintaining the high-availability systems mentioned in the job description.

Your preparation checklist

Prepare a production-grade model deployment case study
Select a project where you moved a model from research to production. Prepare to explain your approach to MLOps, including how you handled latency, containerization with Docker/Kubernetes, and model monitoring in a high-volume environment.
Develop a RAG-based system architecture walkthrough
Create a high-level design document for a RAG-based system that addresses a specific business problem, such as customer intent recognition. Be prepared to discuss your choice of frameworks, vector databases, and how you ensured high-fidelity performance.
Review GCP AI stack and optimization techniques
Refresh your knowledge of Vertex AI and BigQuery. Prepare to discuss how you would optimize data pipelines for real-time travel data, focusing on high-performance coding in Python, Golang, or C++.
Formulate questions on Sabre's AI roadmap
Prepare thoughtful questions regarding the integration of Agentic AI into their travel marketplace, the current challenges in their NDC adoption strategy, or how the team balances classical mathematical optimization with modern Generative AI.
Refine your leadership and mentorship narrative
Prepare specific examples of how you have mentored junior engineers or fostered a culture of rigorous experimentation, as this role requires driving innovation and influencing cross-functional stakeholders.

Availability has not yet been verified. Status can change after a check; confirm on the employer’s website. Report an issue · Our checking policy

KEEP EXPLORING

Similar AI jobs

Related skills and specializations in India. Matched to this role, not your profile.

Explore more jobs

Finding similar opportunities…