New opportunity

Sr. AI Engineer

TELUS Digital · Bengaluru, India

About this role

TELUS Digital is seeking a Sr. AI Engineer to join their team in Bengaluru. The role focuses on building and scaling AI solutions using their enterprise-grade generative AI platform, Fuel iX. Responsibilities include owning architecture for AI systems and ensuring high-quality data pipelines. Requirements: 4–6 years of engineering experience, with at least 2 years specifically focused on shipping production LLM and agentic systems. Candidates must demonstrate depth in agent systems, LLM orchestration, and RAG at scale. Strong data-quality instincts and experience with large-scale pipelines are essential. The role also requires mentorship capabilities and a willingness to work with sensor or media data as needed. Preferred qualifications (Bonus): Experience with multi-agent architectures, inference optimization, ROS/SLAM, and Human-in-the-Loop (HITL) systems.

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

Skills for this role

LLM orchestrationAgentic systemsRAGData engineeringArchitecture designMulti-agent architecturesInference optimizationROSSLAMHITL systems

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

Production-Grade LLM and Agentic System Architecture
The core mandate is to own the architecture for AI systems within the Fuel iX platform. You must demonstrate the ability to move beyond prototypes to shipping and maintaining production-level LLM and agentic workflows, ensuring they are robust, scalable, and reliable for enterprise use.
Data Pipeline Engineering and Quality Assurance
Strong data-quality instincts are required to manage large-scale pipelines. You will be responsible for ensuring the integrity of data feeding into AI models, which is critical for the performance of RAG systems and overall model output accuracy.
Technical Mentorship and Team Leadership
As a Senior AI Engineer, you are expected to provide technical guidance and mentorship to other team members. You should be prepared to discuss your approach to code reviews, architectural decision-making, and fostering technical growth within an engineering team.

Likely challenges to think through

Balancing General LLM Expertise with Domain-Specific Data
The role requires a willingness to work with sensor or media data, which may differ significantly from standard text-based LLM inputs. Prepare by articulating how you adapt RAG or agentic architectures to handle non-traditional data modalities while maintaining system performance.
Optimizing Inference for Real-World Scale
Scaling AI solutions from pilot to production often involves significant latency and cost trade-offs. You may be challenged on your experience with inference optimization techniques and how you balance model complexity against the performance requirements of enterprise-grade applications.

Practice your answers

Technical practice

Designing RAG Systems at Scale

What to coverExplain your approach to building RAG pipelines that handle high throughput while maintaining retrieval accuracy. Discuss how you manage vector database indexing, chunking strategies, and the trade-offs between latency and retrieval quality.

LLM Orchestration Strategies

What to coverDescribe the frameworks or patterns you use for orchestrating LLM workflows. Focus on how you handle state management, error handling, and tool-calling capabilities in production-grade agentic systems.

Data Quality for AI Pipelines

What to coverGiven the requirement for strong data-quality instincts, explain how you validate and clean large-scale datasets before they are used for LLM training or RAG. Suggestion: Prepare examples of how you detect data drift or bias in production pipelines.

Inference Optimization Techniques

What to coverDiscuss methods for reducing latency and cost when deploying LLMs at scale. Address techniques such as quantization, caching strategies, or model distillation that you have implemented in production environments.

Multi-Agent Architecture Design

What to coverExplain the challenges of coordinating multiple specialized agents to complete a complex task. Discuss how you manage inter-agent communication, conflict resolution, and task decomposition.

Integrating Sensor and Media Data

What to coverHow would you approach incorporating non-textual data like sensor or media streams into an AI system? Suggestion: Research how multimodal models or feature extraction pipelines can be integrated into existing LLM architectures.

Human-in-the-Loop (HITL) Implementation

What to coverDescribe how you design systems that incorporate human feedback to improve model performance or safety. Discuss the trade-offs between automated decision-making and human intervention points.

Robotics and Spatial Awareness (ROS/SLAM)

What to coverIf you have experience with ROS or SLAM, explain how these systems interface with AI agents for spatial reasoning. If you lack direct experience, discuss how you would approach learning these frameworks to support AI-driven robotics.

Behavioural practice

Mentoring Junior Engineers

What to coverDescribe a time you mentored a team member to improve their technical skills or output. Focus on how you balanced your own project responsibilities with the need to provide effective guidance.

Owning System Architecture

What to coverProvide an example of a time you were responsible for the end-to-end architecture of a production system. Explain how you made critical design decisions and how you ensured the system remained scalable and maintainable.

Handling Production Failures

What to coverTell me about a time a production AI system failed or performed unexpectedly. Describe your process for diagnosing the root cause, implementing a fix, and preventing future occurrences.

Navigating Ambiguity in AI Projects

What to coverDescribe a situation where you had to build an AI solution with unclear requirements or evolving technology. Explain how you managed stakeholder expectations and iterated on the solution to deliver value.

Sharpen your knowledge

Topics to prepare

LLM Orchestration and Agentic Systems
Review the conceptual framework of agentic workflows, specifically how to manage state, tool-use, and memory in production environments. Exercise: Design a high-level architecture diagram for a multi-agent system that handles a complex customer service query, focusing on how the orchestrator manages task delegation and error handling.
RAG at Scale and Data Quality
Focus on the challenges of scaling Retrieval-Augmented Generation, including vector database performance, retrieval latency, and data pipeline integrity. Exercise: Draft a strategy for evaluating retrieval accuracy and data freshness in a pipeline that processes large volumes of unstructured media or sensor data.
Human-in-the-Loop (HITL) Systems
Understand the integration of human feedback loops into automated AI pipelines to ensure safety and accuracy. Exercise: Propose a mechanism for incorporating expert human feedback into an LLM-based decision-making system, specifically addressing how to handle conflicting feedback and maintain system throughput.

Skills to revise

Inference Optimization
Practice techniques to reduce latency and cost for LLM deployments, such as quantization, caching strategies, and model distillation. Exercise: Analyze a hypothetical production LLM deployment and identify three specific bottlenecks, then propose optimization techniques to improve throughput without sacrificing output quality.
Large-Scale Data Pipeline Engineering
Develop proficiency in building robust, scalable pipelines capable of handling diverse data types including sensor and media inputs. Exercise: Create a schema for a data pipeline that ingests, cleans, and vectorizes raw media files for use in a RAG system, ensuring data quality checks are embedded at each stage.
Multi-Agent Architecture Implementation
Gain hands-on experience with frameworks that support multi-agent collaboration. Exercise: Build a small-scale prototype using a framework like LangGraph or AutoGen where two specialized agents interact to solve a multi-step problem, demonstrating clear communication protocols and task hand-offs.

Your preparation checklist

Prepare a production LLM system case study
Draft a technical walkthrough of a production-grade LLM or agentic system you have shipped. Focus on your architectural decisions, how you handled RAG at scale, and the specific challenges you overcame regarding LLM orchestration.
Document data pipeline architecture
Create a diagram or summary of a large-scale data pipeline you designed. Highlight your approach to ensuring data quality, as this is a core requirement for the Fuel iX platform context.
Develop a strategy for inference optimization
Prepare to discuss methods for optimizing LLM inference, such as quantization, caching strategies, or model distillation, to demonstrate your ability to scale AI solutions efficiently.
Formulate mentorship examples
Prepare specific anecdotes demonstrating your experience mentoring junior engineers, including how you have guided team members through complex technical hurdles or code reviews.
Draft technical questions for the interviewers
Prepare questions regarding the Fuel iX platform architecture, how the team handles multi-agent orchestration in production, and how they integrate sensor or media data into their current AI workflows.

Availability last checked: 18 Oct 2026. 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…