New opportunity

AI Architect

Arrow Electronics · Bangalore, India

About this role

Arrow Electronics seeks an AI Architect in Bangalore to design, develop and implement AI solutions that address business problems and align with organizational goals. The role spans generative AI, machine learning, natural language processing, computer vision and predictive analytics, including development and deployment initiatives for an in-house GenAI platform.

Responsibilities

Define AI architectures, frameworks and strategies; evaluate tools and platforms; and design scalable, secure, maintainable solutions that integrate with existing business processes and comply with Arrow’s internal policies.
Supervise or contribute to developing, training and optimizing large and small language models. Apply on-premises and cloud architecture, including deployment of PaaS and SaaS applications on Microsoft Azure, and work with external vendors and partners.
Set project scope, deliverables, plans, timelines and milestones. Manage daily execution and delivery, work with business units to determine ROI and its measurement KPIs, and address risks involving data limitations, algorithm performance and regulation.
Ensure compliance with data privacy regulations, industry standards and ethical guidelines, and promote responsible AI use.
Coordinate AI researchers, data scientists, engineers, product managers and business stakeholders; communicate progress and challenges between business units and IT; and coordinate onsite and offshore AI teams.

Qualifications

Bachelor’s or Master’s degree in Computer Science, Artificial Intelligence, Data Science or a related field; an advanced degree is preferred. At least 5 years of experience in AI projects, data science or a related field.
Expertise in AI and GenAI methods, including supervised and unsupervised learning, deep learning, NLP and computer vision. Proficiency in Python and .NET; knowledge of Azure, GCP, APIs, microservices, big-data tools such as Hadoop and Apache Spark, and AI development platforms such as Azure AI Foundry, Ollama and Databricks.
Experience architecting on-premises and cloud AI solutions and managing technical projects using agile methods and tools such as Jira or Trello. Strong prioritization, problem-solving, analytical and communication skills; knowledge of AI ethics, regulatory compliance, data engineering and cloud architecture principles.

The position is listed as full-time at the SKAV Seethalakshmi location in Bangalore, India.

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

Skills for this role

AI architectureGenerative AIMachine LearningSupervised learningUnsupervised learningDeep LearningNatural Language ProcessingComputer VisionPredictive analyticsLarge Language ModelsSmall Language ModelsPython.NETMicrosoft AzureGoogle Cloud PlatformAzure AI FoundryOllamaDatabricksHadoopApache SparkAPIsMicroservicesData engineeringCloud architectureAzure PaaSAzure SaaSAgile methodologiesJiraTrelloTechnical project management disciplines not listed

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

Architect AI solutions that fit the business
The role defines AI architectures and selects tools and frameworks while integrating solutions with existing processes. Designs must remain scalable, secure, maintainable and aligned with Arrow’s internal policies.
Advance the in-house GenAI platform
The posting calls for planning development and deployment initiatives for the in-house GenAI platform, and supervising or contributing to the development, training and optimization of large and small language models.
Turn AI work into measurable delivery
The architect sets scope, deliverables, timelines and milestones; manages day-to-day execution; and works with business units to determine ROI and KPIs for each project.
Build governance into execution
The role anticipates data, model-performance and regulatory risks, develops mitigations, and ensures projects follow data privacy requirements, industry standards and ethical guidelines.
Coordinate technical and business teams
The architect serves as a liaison across AI research, data science, engineering, product and business stakeholders, including coordination between onsite and offshore AI teams.

Likely challenges to think through

Balance deployment choices with operational constraints
A likely design trade-off is choosing among on-premises, cloud and vendor-supported components without sacrificing integration, security or maintainability. Prepare to explain how you would compare options, including Azure PaaS or SaaS deployment, against those constraints.
Commit to outcomes amid model uncertainty
AI project plans may need firm milestones and ROI KPIs even when data limitations or model performance are uncertain. Prepare an example approach to setting measurable checkpoints, testing assumptions and defining contingencies before committing to broader deployment.
Keep parallel teams and projects aligned
Managing multiple initiatives across business units, IT, vendors, and onsite and offshore teams could create competing priorities or handoff delays. Prepare to describe how you would make ownership, dependencies, decisions and progress visible while adjusting plans when risks emerge.

Practice your answers

Technical practice

How would you design an AI solution that fits an existing business process?

What to coverSuggested answer focus: Explain how you would translate business objectives into architecture and integration choices, including where APIs, microservices, Python or .NET might fit. Discuss scalability, security, maintainability and compliance with internal policies.

How would you choose between on-premises and cloud deployment for an AI solution?

What to coverSuggested answer focus: Compare data constraints, security, scalability, operations and cost. Address how Azure PaaS or SaaS deployment could fit the design, while noting when another cloud or on-premises approach may be preferable.

How would you develop and optimize a large or small language model for a business need?

What to coverSuggested answer focus: Describe how you would define the task, select an appropriate model size and evaluate quality before deployment. Discuss trade-offs among performance, latency, resource use and ongoing maintenance.

How would you select an AI method for a proposed business problem?

What to coverSuggested answer focus: Explain what would lead you toward supervised or unsupervised learning, deep learning, NLP, computer vision or predictive analytics. Tie the choice to available data, the intended outcome and a measurable evaluation approach.

How would you prepare data for an AI solution at scale?

What to coverSuggested answer focus: Cover data quality, access and processing needs, and explain when tools such as Spark or Hadoop would be useful. Discuss how data limitations could affect model design and delivery plans.

How would you evaluate AI development platforms and external solutions?

What to coverSuggested answer focus: Propose criteria for comparing tools such as Azure AI Foundry, Ollama or Databricks against the use case and existing architecture. Include integration, security, maintainability and the trade-off between building internally and working with a vendor.

What would you do if an AI solution performed poorly during evaluation?

What to coverSuggested answer focus: Describe how you would separate data limitations from algorithm or implementation issues and choose meaningful performance measures. Explain how the findings would inform optimization, a change of approach or a project risk decision.

How would you incorporate privacy and responsible AI into a solution design?

What to coverSuggested answer focus: Explain how you would identify applicable privacy, industry, ethical and internal-policy constraints before deployment. Discuss practical design and review decisions without assuming a particular regulation or company procedure.

How would you define ROI measures for an AI project?

What to coverSuggested answer focus: Describe how you would agree on a business objective and KPIs with the relevant business unit. Explain how you would establish a baseline and distinguish technical model performance from measurable business value.

How would you plan delivery for an in-house GenAI platform initiative?

What to coverSuggested answer focus: Break development and deployment into scope, deliverables, milestones and risks, with room to adjust as results emerge. Explain how agile tracking could make dependencies and progress visible without treating a named tool as mandatory.

Behavioural practice

Tell me about a time you aligned technical and business stakeholders on an AI project.

What to coverSuggested answer focus: Use a real situation to show the differing objectives, the actions you took to clarify scope or trade-offs, and the result. Make your own contribution to communication between business and technical groups clear.

Describe a time you coordinated work across teams or locations.

What to coverSuggested answer focus: Give the situation, how you organized responsibilities and communication, and the outcome. An example involving onsite and offshore teams is relevant if you have one; do not imply that experience if you do not.

Tell me about a time you had to prioritize competing technical projects.

What to coverSuggested answer focus: Set out the competing demands, the criteria you used to prioritize them and the result. Explain how you communicated decisions and managed effects on timelines or deliverables.

Describe a project risk you identified before it disrupted delivery.

What to coverSuggested answer focus: Use a real example involving data, performance, compliance or another relevant risk. Explain the situation, your mitigation or contingency actions, and what happened.

Tell me about a time you removed an obstacle to collaboration.

What to coverSuggested answer focus: Describe the obstacle, the people affected, your actions and the result. Focus on how you helped researchers, engineers, product colleagues or other stakeholders make progress rather than claiming credit for the whole team.

Sharpen your knowledge

Topics to prepare

AI architecture and integration
The posting calls for scalable, secure, maintainable AI solutions that fit existing business processes and internal policies. As a revision exercise, sketch an architecture for a hypothetical business problem and explain its integration points, security choices and maintenance trade-offs.
Model and method selection
The role spans GenAI, supervised and unsupervised learning, deep learning, NLP, computer vision and predictive analytics, including large and small language models. As a revision exercise, compare which approaches suit three hypothetical problems and explain how you would evaluate and optimize each choice.
On-premises and cloud architecture
The posting specifies on-premises and cloud AI architecture, Azure PaaS and SaaS deployment, and knowledge of GCP. As a revision exercise, make a decision matrix comparing on-premises and cloud options for scale, integration, security and maintainability, with particular attention to Azure deployment.
Responsible AI and project risk
The role requires attention to data privacy, industry standards, ethical guidelines, data limitations, algorithm performance and regulatory concerns. As a revision exercise, write a risk register for a hypothetical AI initiative, pairing each risk with a mitigation and a way to monitor it; do not assume a particular regulation applies.
Business value and delivery planning
The architect defines scope, milestones and deliverables with business units, including ROI and measurement KPIs for the in-house GenAI platform. As a revision exercise, turn a hypothetical business objective into a project scope, measurable KPIs, a milestone plan and an explanation of how results would inform ROI.

Skills to revise

Prototype and assess language models
The posting includes contributing to the development, training and optimization of large and small language models. As a practice exercise, compare two model sizes on a small sample task, record quality and resource trade-offs, and explain what you would change before deployment.
Build an AI integration interface
Python, .NET, APIs and microservices are named in the qualifications. As a practice exercise, implement a small AI-facing API in Python or .NET and describe how another business application would call it and handle errors.
Work through an Azure deployment design
The role explicitly calls for deploying PaaS and SaaS applications on Microsoft Azure. As a practice exercise, outline deployment steps and a configuration checklist for a hypothetical AI application, including access, integration, scaling and operational checks.
Evaluate named AI platforms
The posting names Azure AI Foundry, Ollama and Databricks as examples of AI development platforms. As a practice exercise, trial or review two of them against the same hypothetical need and document selection criteria and limitations; this comparison is preparation, not a stated project requirement.
Prepare data for AI workloads
Data engineering principles and big-data tools such as Hadoop and Apache Spark appear in the qualifications. As a practice exercise, use Spark to clean and aggregate a sample dataset, then document data-quality limitations that could affect model performance.
Coordinate an agile AI delivery plan
The role involves daily execution, prioritization and coordination across business, IT, onsite and offshore teams; Jira and Trello are listed as example tools. As a practice exercise, create a short backlog and milestone board in either tool, then draft a concise stakeholder update covering progress, dependencies and risks.

Your preparation checklist

Prepare an AI project walkthrough
Choose a relevant project and prepare a brief account of the business problem, your architecture decisions, your contribution, deployment, measured results and trade-offs. The role calls for AI solutions aligned with business objectives.
Practice an in-house GenAI platform design exercise
Sketch a hypothetical solution with data flow, model choices, APIs, integration with existing processes, and on-premises versus cloud deployment. Explain how you would make it scalable, secure and maintainable; do not assume Arrow’s current platform design.
Collect evidence of technical depth
Prepare specific examples of Python and .NET work, Azure deployments, and any hands-on use of tools named in the posting, such as Azure AI Foundry, Ollama or Databricks. Be ready to explain what you personally built and why you chose each tool.
Explain model development decisions
Prepare an example of developing, training or optimizing a large or small language model, including evaluation and performance limitations. The posting also names ML, NLP, computer vision and predictive analytics; focus on areas where you have concrete evidence.
Draft a delivery and ROI plan
For a hypothetical AI initiative, outline scope, milestones, dependencies, risks and contingency actions. Add proposed KPIs and a way to work with a business unit to assess ROI, as the role includes project delivery and ROI measurement.
Prepare governance and coordination examples
Identify examples of addressing privacy, regulatory or ethical risks and of coordinating technical and business stakeholders. Prepare to explain how you communicated trade-offs and organized work across teams, including onsite and offshore teams where relevant.
Write questions for the employer
Ask which in-house GenAI platform priorities matter most, how architecture decisions and internal-policy reviews are handled, what ROI measures business units use, and how the AI teams and external partners divide responsibilities.

Availability last checked: 5 Nov 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…