Senior Data Science Engineer

Sabre; Bengaluru; On-site; INR 3,000,000 - 6,000,000; Full time

Job summary

Sabre in Bengaluru seeks a Senior Engineer to design and implement production-grade data pipelines on Google Cloud using Dataflow/Beam, Vertex AI, GenAI and ADK-based agentic AI solutions. You will ensure reliability, safety, and cost efficiency while mentoring juniors and building reusable patterns for lakehouse loading.

You will collaborate with leaders to enforce governance, security, and observable systems, delivering high-quality code with strong testing and documentation.

Qualifications

• 5-8 years in software/data/ML engineering with GenAI/agentic systems.
• Hands-on with GCP AI stack and ADK-based agent development.
• Strong coding in Python/TypeScript; infra-as-code familiarity.
• Experience with Java, Dataflow + Beam or Spark; LLMOps in production.

Responsibilities

• Design and implement GenAI workflows, RAG pipelines, embeddings, and evaluation.
• Build agentic AI components: planners, tools, memory, and guardrails using ADK.
• Integrate Vertex AI, BigQuery, Cloud Storage, Pub/Sub, Cloud Run, Workflows.
• Ensure observability, testing, and cost optimization; enable reliable operation.
• Mentor juniors; participate in design reviews, code reviews, and knowledge sharing.
• Apply security controls (IAM, VPC-SC) and Responsible AI governance.

Skills

GenAI

Agentic AI

GCP

LLMOps

Data pipelines

Java

Python

TypeScript

Tools

Vertex AI

BigQuery

Dataflow/Beam

Cloud Storage

Pub/Sub

ADK

Job description

Role Summary

The Senior Engineer is hands‑on technical expert responsible for designing and implementing data pipeline using dataflow and Beam, GenAI and Agentic AI solutions on Google Cloud Platform using Vertex AI and ADK frameworks. This role focuses on building production‑grade systems, ensuring reliability, safety, and cost efficiency, and mentoring junior engineers while contributing to reusable patterns and best practices. The engineer should also develop data pipelines to load into lakehouse.

Role Summary

The Senior Engineer is hands‑on technical expert responsible for designing and implementing data pipeline using dataflow and Beam, GenAI and Agentic AI solutions on Google Cloud Platform using Vertex AI and ADK frameworks. This role focuses on building production‑grade systems, ensuring reliability, safety, and cost efficiency, and mentoring junior engineers while contributing to reusable patterns and best practices. The engineer should also develop data pipelines to load into lakehouse.

Key Responsibilities

Solution Design & Development

• Implement GenAI workflows: prompt engineering, RAG pipelines, embeddings, and evaluation.
• Build agentic AI components: planners, tools, memory management, and guardrails using ADK.
• Integrate GCP services: Vertex AI, BigQuery (including vector functions), Cloud Storage, Pub/Sub, Cloud Run, Workflows.

Delivery & Quality

• Write clean, maintainable code with proper documentation and testing.
• Ensure operational readiness: observability, logging, error handling, retries, and rollback mechanisms.
• Optimize performance and cost through caching, batching, and adaptive routing.

Collaboration & Mentorship

• Work closely with Team Lead and Principal Engineer to align on architecture and standards.
• Mentor junior engineers on prompt engineering, agent design, and GCP best practices.
• Participate in code reviews, design discussions, and knowledge‑sharing sessions.

Governance & Compliance

• Implement security controls: IAM, VPC‑SC, Secret Manager, and data residency requirements.
• Apply Responsible AI principles: safety prompts, content filters, and audit logging.

Required Technical Competencies

• GenAI: Prompt engineering, RAG, embeddings, fine‑tuning, evaluation metrics.
• Agentic AI (ADK): Agent loops, tool integration, memory handling, planning strategies.
• GCP Services: Vertex AI, BigQuery, Cloud Storage, Pub/Sub, Cloud Run, Workflows.
• LLMOps: CI/CD pipelines, model registry, telemetry, cost/performance dashboards.
• Security & Compliance: IAM, VPC‑SC, DLP, Okta/IAP integration.
• Data pipeline: Dataflow, Apache Beam, Java.

Qualifications

• 5-8 years in software/data/ML engineering; 1-2 years in GenAI/agentic systems.
• Hands‑on experience with GCP AI stack and ADK‑based agent development.
• Strong coding skills in Python/TypeScript and familiarity with infrastructure‑as‑code.
• Hands‑on experience on Java, Dataflow + Beam or Spark.
• Exposure to LLMOps practices and production deployments.

Outcomes & KPIs

• Delivery: Features shipped on time with minimal defects.
• Quality: Evaluation metrics (faithfulness, grounding) meet thresholds.
• Cost: Demonstrates cost optimization in design and implementation.
• Team Contribution: Active participation in reviews, documentation, and mentoring.

Demonstrated Behaviors (Senior Engineer Level)

Technical Execution

• Delivers high‑quality, tested code aligned with architecture standards.
• Proactively identifies performance and reliability improvements.

Collaboration

• Works effectively with team members; shares knowledge openly.
• Communicates risks and blockers early; seeks help when needed.

Responsible AI

• Applies safety and compliance measures consistently.
• Raises concerns about ethical or security risks promptly.