New opportunity

Associate Architect - Machine Learning (Azure)

Quantiphi · Bangalore, India

About this role

Quantiphi is seeking an innovative Associate Architect - Machine Learning (Azure) to lead the design, development, and deployment of high-impact, data-driven solutions. This role focuses on architecting complex systems, including multi-agent platforms and computer vision solutions, while providing technical leadership to cross-functional teams. Responsibilities: Lead the end-to-end ML lifecycle, including data preparation, feature engineering, model development, validation, deployment, and monitoring. Architect complex multi-agent systems using patterns like supervisor-worker and orchestrator-led group chats. Design and implement RAG pipelines on Azure, integrating LLMs with domain-specific knowledge bases. Optimize legacy models using traditional ML algorithms and ensure adherence to MLOps best practices. Design fault-tolerant systems with robust error handling and state management for complex AI workflows. Provide technical mentorship and architecture-level guidance to project teams. Requirements: 8-14 years of experience in architecting and deploying traditional and deep learning solutions at scale. Extensive experience with Azure cloud services (Azure ML, Data Factory, Blob Storage, DevOps, Container Apps). Proven expertise in traditional ML algorithms (regression, decision trees, SVM, ensemble models, clustering, Random Forest, XGBoost). Experience with data augmentation and human-in-the-loop annotation processes. Strong problem-solving, communication, and stakeholder management skills. Preferred: Certifications in Azure AI Engineer or Azure Solutions Architect. Experience with evaluation and observability frameworks for Agentic workflows. Contributions to open-source projects. The role is based in Bangalore and operates within a hybrid work culture.

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

Skills for this role

Machine LearningGenerative AIAzure Machine LearningAzure Data FactoryAzure OpenAI ServiceRAGMulti-agent systemsComputer VisionMLOpsPythonRegressionDecision TreesSVMEnsemble modelsClusteringRandom ForestXGBoostAzure DevOpsAzure Container AppsAzure Cognitive SearchSystem Architecture

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

Architecting Complex Multi-Agent Systems
The role requires designing sophisticated AI workflows using patterns like supervisor-worker and orchestrator-led group chats. You must demonstrate how to structure these systems to automate intricate business processes while ensuring scalability and reliability.
End-to-End MLOps and Azure Integration
You are expected to lead the full ML lifecycle on Azure, utilizing services like Azure Machine Learning, Data Factory, and Container Apps. Prioritize demonstrating your ability to implement robust MLOps practices, including model monitoring, deployment, and pipeline orchestration.
RAG Pipeline Implementation
A core duty is building RAG pipelines that integrate LLMs with domain-specific knowledge bases. Focus on your experience with Azure Cognitive Search and Azure OpenAI to ensure accurate, context-aware information retrieval.
Technical Leadership and Mentorship
As an Associate Architect, you must provide guidance to cross-functional teams. Be prepared to discuss your approach to reviewing ML architectures, mentoring junior staff, and translating complex business requirements into actionable technical strategies.

Likely challenges to think through

Balancing Legacy Optimization with Generative AI
You will be tasked with both optimizing traditional ML models and deploying modern Generative AI solutions. The challenge lies in maintaining legacy systems while integrating cutting-edge technology; prepare by highlighting your methodology for evaluating when to refactor versus when to replace existing models.
Ensuring Fault Tolerance in Agentic Workflows
Complex AI workflows involving multiple agents are prone to cascading failures. You will need to design robust error handling and state management; prepare by discussing specific strategies for implementing fallback mechanisms and observability in non-deterministic agentic systems.
Stakeholder Management in Technical Projects
The role requires presenting complex technical solutions to business stakeholders. The difficulty is bridging the gap between high-level business goals and deep technical implementation; prepare by practicing how to articulate the ROI and architectural trade-offs of your AI designs to non-technical audiences.

Practice your answers

Technical practice

Architecting Multi-Agent Systems

What to coverExplain how you would design a supervisor-worker or orchestrator-led agentic pattern. Focus on how you manage state, inter-agent communication, and error handling in complex, multi-step workflows.

RAG Pipeline Implementation on Azure

What to coverDescribe your approach to building a RAG pipeline using Azure OpenAI and Azure Cognitive Search. Discuss strategies for chunking, embedding, and ensuring retrieval accuracy for domain-specific knowledge bases.

MLOps and Lifecycle Management

What to coverDetail your experience implementing MLOps best practices within the Azure ecosystem. Address how you handle model versioning, automated retraining, and monitoring in production environments using Azure DevOps and Azure ML.

Optimizing Traditional ML Models

What to coverWhen tasked with improving legacy models, what is your process for evaluating performance? Discuss how you apply techniques like feature engineering, ensemble methods, or hyperparameter tuning to improve existing regression or classification models.

Fault-Tolerant AI System Design

What to coverHow do you ensure system reliability for AI workflows? Explain your strategies for implementing fallback mechanisms, robust error handling, and state management when dealing with non-deterministic LLM outputs.

Observability for Agentic Workflows

What to coverWhat frameworks or metrics do you use to evaluate and monitor agentic systems? Suggest how you would track agent performance, latency, and cost while maintaining visibility into the decision-making process.

Data Augmentation and Human-in-the-Loop

What to coverDescribe a scenario where you implemented human-in-the-loop annotation for model training. Explain how you balanced data quality requirements with the need for scalable, automated training pipelines.

Azure Infrastructure Integration

What to coverHow do you leverage Azure Container Apps and Data Factory to support scalable ML deployments? Discuss the trade-offs between different compute options for hosting deep learning models versus lightweight inference services.

Behavioural practice

Technical Leadership and Mentorship

What to coverDescribe a time you provided architecture-level guidance to a cross-functional team. Focus on how you communicated complex technical trade-offs to stakeholders and mentored junior team members to achieve project goals.

Stakeholder Management

What to coverShare an example of presenting a technical ML solution to non-technical business stakeholders. How did you ensure your proposal aligned with business requirements while managing expectations regarding technical feasibility?

Collaborating Across Teams

What to coverGive an example of working with data scientists and data engineers to convert a business problem into a scalable ML solution. Highlight how you navigated conflicting priorities or technical constraints.

Handling Project Ambiguity

What to coverDescribe a situation where you had to lead the design of a complex system with evolving requirements. How did you maintain architectural integrity while ensuring the team remained agile and productive?

Sharpen your knowledge

Topics to prepare

Multi-Agent System Architecture
The role requires designing systems using supervisor-worker and orchestrator-led patterns. Exercise: Draft a high-level architecture diagram for an automated document processing system using a multi-agent approach, focusing on how agents communicate and handle state. This is directly relevant to the requirement for automating complex business processes.
RAG Pipeline Design on Azure
The position involves integrating LLMs with domain-specific knowledge bases. Exercise: Outline a robust RAG architecture using Azure Cognitive Search and Azure OpenAI, specifically addressing how you would handle data ingestion, chunking strategies, and retrieval evaluation. This aligns with the need for scalable knowledge retrieval systems.
MLOps and Lifecycle Management
The role demands end-to-end ML lifecycle management and production-grade implementation. Exercise: Define a CI/CD pipeline strategy for an ML model using Azure DevOps and Azure Container Apps, focusing on automated monitoring and model retraining triggers. This is essential for the required adherence to MLOps best practices.

Skills to revise

Azure Cloud Services Integration
Extensive experience with Azure ML, Data Factory, and Blob Storage is required. Exercise: Create a mock project plan that integrates these services to move data from a raw source to a deployed model endpoint. This demonstrates your ability to architect solutions within the specific Azure ecosystem requested.
Fault-Tolerant AI Workflow Design
The role requires designing systems with robust error handling and fallback mechanisms. Exercise: Review a multi-step AI workflow and identify potential failure points, then propose specific fallback strategies or state management techniques to ensure system reliability. This is critical for the requested fault-tolerant system design.
Traditional ML Algorithm Optimization
The role requires evaluating and optimizing legacy models using traditional algorithms. Exercise: Select a common business problem (e.g., churn prediction) and compare the performance and interpretability of a Random Forest versus an XGBoost model, documenting the optimization process. This addresses the requirement for expertise in traditional ML algorithms.

Your preparation checklist

Map Azure ML and MLOps expertise to production scenarios
Prepare to discuss your experience architecting end-to-end ML lifecycles on Azure. Focus on how you have utilized Azure Machine Learning, Data Factory, and DevOps to manage model deployment, monitoring, and pipeline orchestration at scale.
Develop a multi-agent system design walkthrough
Create a technical summary of a multi-agent system you have designed. Be prepared to explain your use of patterns like supervisor-worker or orchestrator-led group chats, specifically addressing how you handled state management and fault tolerance.
Prepare a RAG pipeline architecture case study
Draft a detailed explanation of a RAG pipeline you implemented using Azure OpenAI and Azure Cognitive Search. Highlight how you integrated domain-specific knowledge bases and ensured retrieval accuracy for complex AI workflows.
Synthesize traditional ML and deep learning experience
Review your history with traditional algorithms like XGBoost, Random Forest, and SVM alongside deep learning. Prepare examples of how you have evaluated and optimized legacy models or utilized data augmentation to improve performance.
Formulate questions on Quantiphi's AI engineering culture
Prepare thoughtful questions regarding their approach to observability frameworks for agentic workflows or how they balance technical mentorship with high-impact delivery in their hybrid work environment.

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…