Job Description

Dreaming big is in our DNA. It’s who we are as a company. It’s our culture. It’s our heritage. And more than ever, it’s our future. A future where we’re always looking forward. Always serving up new ways to meet life’s moments. A future where we keep dreaming bigger. We look for people with passion, talent, and curiosity, and provide them with the teammates, resources and opportunities to unleash their full potential. The power we create together – when we combine your strengths with ours – is unstoppable. Are you ready to join a team that dreams as big as you do?

Job Description

Job Title: Senior Data Scientist

Location: Bangalore

Reporting to: Senior Manager - Analytics

PURPOSE OF ROLE

The Senior Data Scientist will design, build, and scale data science and connected agentic systems that improve decision-making and automate high-value business processes across the organization. The role combines strong hands-on software and machine learning engineering with the ability to translate ambiguous business problems into reliable, production-ready solutions. The successful candidate will have demonstrable experience building agentic projects, including systems in which multiple specialized agents, tools, data sources, and human controls work together to complete an end-to-end workflow. They will bring technical leadership to architecture, evaluation, deployment, and adoption while partnering with business and technology stakeholders to deliver measurable impact.

KEY TASKS AND ACCOUNTABILITIES

• Lead the end-to-end design and delivery of connected agentic systems that combine LLMs, specialized agents, tools, APIs, structured and unstructured data, retrieval, memory/state, and human-in-the-loop controls.
• Own the architecture of multi-agent workflows, including agent roles, orchestration, task decomposition, routing, tool use, hand-offs, error recovery, guardrails, and escalation paths.
• Build and productionize data science and machine learning solutions using Python and modern libraries, applying sound statistical validation, feature engineering, and model evaluation practices.
• Develop robust RAG and knowledge-grounding patterns, including ingestion, chunking, embeddings, retrieval, reranking, citation/traceability, freshness, and access-control considerations.
• Create evaluation frameworks for agentic systems covering task success, factuality, groundedness, safety, latency, cost, reliability, and business outcomes; establish monitoring and continuous-improvement loops.
• Translate business needs into product and technical requirements, prioritize use cases, define success metrics, and communicate trade-offs clearly to senior stakeholders.
• Write clean, modular, testable, and well-documented code; use Git, code reviews, CI/CD, testing, observability, and reproducible development practices to support reliable delivery.
• Read, understand, and extend existing codebases and platforms, adapting quickly to different coding styles, project structures, data environments, and technology constraints.
• Write and optimize SQL and data pipelines to extract, transform, validate, and analyze data from structured and semi-structured sources.
• Provide technical mentorship to data scientists and engineers, raise engineering standards, and contribute reusable patterns, reference architectures, and best practices across teams.
• Document system architecture, prompts, tools, assumptions, risks, decisions, operating procedures, and model/system limitations to support transparency and responsible adoption.

AGENTIC SYSTEM EXPERIENCE

This is a core requirement. Candidates must be able to demonstrate, with specific examples, their individual contribution to building and delivering connected agentic systems.

• Demonstrable hands-on experience building at least one agentic project beyond a simple chatbot or prompt wrapper; examples may include multi-agent orchestration, tool-using agents, autonomous or semi-autonomous workflows, agentic RAG, or agents integrated with enterprise systems.
• Ability to explain the architecture, agent responsibilities, control flow, tools/data connections, failure modes, evaluation approach, security considerations, deployment model, and measurable outcome of an agentic project.
• Evidence of production or pilot delivery, such as a deployed system, working prototype with users, code repository, architecture document, technical demo, portfolio, or other artifacts that validate the candidate’s contribution. Candidates should clearly distinguish their own work from team-level results.

QUALIFICATIONS

• Bachelor’s or master’s degree in Computer Science, Information Systems, Artificial Intelligence, Machine Learning, Data Science, Engineering, or a related quantitative field (B. Tech / BE / Masters in CS/IS/AI/ML/DS or equivalent).

Work Experience

• Minimum of 6 years of professional experience in data science, machine learning, applied AI, software engineering, or a closely related role, including at least 2 years building and delivering LLM, generative AI, or agentic AI solutions. Equivalent depth of experience may be considered for exceptional candidates.

Technical Skills Required

Python Programming (Advanced)

Strong command of Python, data structures, object-oriented design, APIs, asynchronous/concurrent patterns where relevant, and production-quality engineering practices.

Agentic AI and LLM Engineering

Strong understanding of agent design patterns, orchestration, tool/function calling, structured outputs, prompt and context engineering, planning/reasoning patterns, memory/state, multi-agent coordination, and human-in-the-loop controls.

RAG and Knowledge Systems

Hands-on experience with vector search, embeddings, retrieval pipelines, grounding, document/knowledge ingestion, metadata, access controls, and strategies to improve answer quality and traceability.

Evaluation, Safety, and Observability

Ability to define and implement evaluations for quality, reliability, safety, latency, and cost; experience with tracing, monitoring, guardrails, red-teaming, fallback behavior, and responsible AI practices.

Data Science Fundamentals

Strong statistical reasoning, experiment design, feature engineering, model evaluation, and ability to connect analytical rigor to business decisions.

SQL and Data Engineering

Proficiency in writing and optimizing SQL and working with data pipelines, APIs, structured/semi-structured data, and cloud or enterprise data platforms.

Software Delivery and Version Control

Experience with Git, code reviews, testing, CI/CD, containerization and/or cloud deployment, reproducibility, documentation, and operating production systems.

Stakeholder and Technical Leadership

Ability to lead through influence, mentor others, communicate technical concepts to non-technical stakeholders, and make pragmatic architecture and delivery trade-offs.

GOOD TO HAVE

• AI product experience: Experience taking an AI/ML capability from problem discovery and user research through prototyping, prioritization, launch, adoption, and iteration.
• Process transformation: Track record of redesigning or automating an end-to-end business process using agentic systems, with measurable improvements in cycle time, quality, productivity, cost, or user experience.
• Experience with agentic frameworks and platforms such as LangGraph, LangChain, Semantic Kernel, AutoGen, CrewAI, or comparable orchestration approaches; framework choice should be supported by sound architectural reasoning.
• Experience with cloud AI services, model gateways, model selection, fine-tuning or adaptation, synthetic data, and cost/performance optimization.
• Knowledge of MLOps/LLMOps, data governance, privacy, security, identity/access management, and enterprise risk controls for AI systems.
• Test-Driven Development, automated evaluation, and strong technical writing/documentation practices.
• Language: English required; Portuguese and/or Spanish are an advantage.

The candidate should be comfortable operating in an evolving technology landscape, making principled decisions under ambiguity, and balancing experimentation speed with production reliability and responsible AI.