New opportunity

Principal Engineer- Local AI/Hardware AI

Insight · Gurugram, India

About this role

As a Principal Engineer/Architect at Insight, you will lead the adoption of local AI deployment and agentic workflows for enterprise clients across edge and desktop AI hardware environments. You will be responsible for unpacking complex business challenges, defining practical AI use cases, building proofs of concept, and enabling stakeholders to scale internal AI solutions. Responsibilities include partnering with enterprise stakeholders to map technical requirements and develop business narratives, implementing and optimizing local AI environments on NVIDIA GB10 and Apple Silicon systems, and building functional AI prototypes including command-line tools and agentic workflows. You will also configure and troubleshoot Linux environments, local models, and networking components. Requirements include working familiarity with running local models on NVIDIA GB10 and Apple Silicon hardware, proficiency with developer tools like Ollama, vLLM, and LM Studio, and a solid understanding of core LLM mechanics, prompt design, and agent patterns. Candidates must be comfortable navigating Linux terminal environments, writing Bash/Python automation scripts, and managing system dependencies. A strong understanding of networking fundamentals such as TCP/IP, DNS, RESTful APIs, and webhooks is required. This is a hybrid role based in India (Delhi NCR, Hyderabad, Bangalore, Pune) with a US-aligned night shift schedule. Benefits include medical insurance, health benefits, professional development support, and a flexible work policy allowing for 30 days of remote work from other locations annually.

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

Skills for this role

Local AI deploymentAgentic workflowsEdge AIDesktop AI hardwareNVIDIA GB10Grace BlackwellCUDAApple SiliconMLXOllamavLLMLM StudioLLM mechanicsPrompt designLinuxBashPythonTCP/IPDNSRESTful APIsWebhooks

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

Local AI Deployment and Optimization
The role requires hands-on implementation of local AI environments on specific hardware, including NVIDIA GB10 and Apple Silicon. You must demonstrate expertise in managing system dependencies, optimizing for latency and cost, and utilizing tools like vLLM, Ollama, and LM Studio to ensure high-performance local model execution.
Agentic Workflow Development
You are expected to build functional AI prototypes and agentic workflows that solve real-world business problems. This involves designing effective prompt strategies, implementing function-calling solutions, and creating automation scripts that enable enterprise clients to scale their internal AI capabilities.
Stakeholder Engagement and Technical Narrative
As a Principal Engineer, you must bridge the gap between complex technical requirements and business outcomes. You will partner with enterprise stakeholders to identify bottlenecks, define high-impact use cases, and articulate clear technical narratives to secure executive buy-in for AI adoption.
Infrastructure and Networking Integration
Successful delivery requires deep knowledge of Linux environments and networking fundamentals. You must be proficient in configuring RESTful APIs, webhooks, and secure network traffic, ensuring that local AI deployments integrate seamlessly into existing client infrastructure.

Likely challenges to think through

Hardware-Specific Constraints
Optimizing models for both NVIDIA GB10 and Apple Silicon requires managing distinct memory architectures and frameworks like CUDA and MLX. Prepare by discussing how you handle hardware-specific performance tuning and resource management to maintain stability in edge environments.
US-Aligned Shift Coordination
The role operates on a US-aligned night shift schedule in India. This can create communication gaps with local teams or stakeholders. Prepare by highlighting your experience with asynchronous workflows and strategies for maintaining high-quality collaboration while working across time zones.
Rapidly Evolving AI Ecosystem
The field of local AI and agentic frameworks changes quickly. You will be challenged to maintain technical relevance while delivering immediate client value. Prepare by sharing your methodology for continuous learning and how you evaluate new frameworks before recommending them for enterprise use.

Practice your answers

Technical practice

Optimizing Local AI on NVIDIA GB10 vs. Apple Silicon

What to coverDiscuss the architectural differences between NVIDIA Grace Blackwell and Apple Silicon when deploying local LLMs. Focus on how you manage memory constraints, such as unified memory on Apple hardware versus dedicated VRAM on NVIDIA, and how these impact model selection and inference latency.

Implementing Agentic Workflows

What to coverExplain your approach to designing agentic workflows that utilize function calling. Describe how you ensure reliability when an LLM interacts with external tools or APIs, and how you handle error states or unexpected model outputs within an automated pipeline.

Managing Local AI Environments

What to coverDescribe your process for setting up and maintaining local AI environments using tools like Ollama, vLLM, or LM Studio. Address how you manage system dependencies, environment isolation, and version control for models in a production-adjacent enterprise setting.

Networking and Security for Local AI

What to coverExplain how you secure local AI deployments that require external connectivity via RESTful APIs or webhooks. Discuss best practices for managing local port forwarding, DNS configuration, and ensuring secure data handling when integrating local models with enterprise systems.

Automation with Bash and Python

What to coverProvide examples of how you use Bash and Python to automate the deployment or monitoring of AI infrastructure. Focus on tasks such as system configuration, dependency management, or creating command-line tools that simplify the interaction between users and local models.

Prompt Design and LLM Mechanics

What to coverDiscuss your methodology for prompt engineering when building prototypes for enterprise use cases. Explain how you balance model performance, token usage, and output consistency, and how you iterate on prompts based on feedback from stakeholders.

Troubleshooting Linux-based AI Systems

What to coverWalk through a scenario where a local AI service is failing to respond to API requests. Detail your systematic approach to debugging the issue, covering potential points of failure such as network connectivity, resource exhaustion, or model-specific configuration errors.

Behavioural practice

Translating Business Challenges into AI Solutions

What to coverDescribe a time you worked with a non-technical stakeholder to identify a business bottleneck. Explain how you mapped their requirements to a technical AI solution and how you communicated the value of the prototype to gain their buy-in.

Managing US-Aligned Shift and Collaboration

What to coverThis role requires working a US-aligned night shift. Share your experience or strategy for maintaining high-quality collaboration and communication with global teams when working across different time zones.

Rapid Self-Directed Learning

What to coverThe AI field evolves quickly; provide an example of a new framework or tool you recently learned on your own. Explain how you integrated this knowledge into your work and how you stay current with emerging AI technologies.

Coaching and Enabling Stakeholders

What to coverGive an example of a time you had to coach a client or teammate through the adoption of a new technology. How did you simplify complex technical concepts to ensure they felt confident using the new AI solution?

Sharpen your knowledge

Topics to prepare

Agentic Workflow Patterns
Review foundational agentic patterns such as ReAct, tool-use, and function calling. Exercise: Design a conceptual architecture for an autonomous agent that uses a local LLM to interact with a REST API to perform a specific business task. This is relevant to the requirement of building functional agentic prototypes for enterprise clients.
Local LLM Deployment Mechanics
Study the differences between running models on NVIDIA Grace Blackwell (CUDA-based) versus Apple Silicon (MLX/Unified Memory). Exercise: Compare the configuration steps for optimizing a model for low-latency inference on both hardware architectures. This is directly relevant to the role's focus on optimizing local AI environments.
Enterprise Networking Fundamentals
Refresh knowledge on TCP/IP, DNS, and secure network traffic, specifically regarding local port forwarding and webhooks. Exercise: Create a diagram showing how a local AI model would securely expose a service via a webhook to an external enterprise application. This is essential for the troubleshooting and integration responsibilities mentioned in the job description.

Skills to revise

Local AI Tooling Proficiency
Gain hands-on experience with Ollama, vLLM, and LM Studio. Exercise: Deploy a quantized model using each tool and measure the resource utilization and inference latency on your local machine. This is a core requirement for enabling local AI deployment for clients.
Bash and Python Automation
Practice writing scripts to manage system dependencies and automate model deployment tasks. Exercise: Write a Python script that uses a library to interact with a local model's API and a Bash script to monitor system resource usage during inference. This is required for managing the local AI environments described in the role.
Linux Environment Management
Strengthen skills in navigating and troubleshooting Linux terminal environments. Exercise: Configure a Linux environment to run a local model as a background service, ensuring proper handling of environment variables and system dependencies. This is necessary for the configuration and troubleshooting tasks listed in the job description.

Your preparation checklist

Develop a Local AI Prototype
Build a functional proof-of-concept using Ollama or vLLM on a local machine. Demonstrate an agentic workflow that utilizes function calling to interact with a RESTful API, showcasing your ability to bridge local model inference with external data sources.
Prepare a Technical Case Study on Hardware Optimization
Create a brief presentation or document detailing how you would optimize model performance on specific hardware like Apple Silicon (using MLX) or NVIDIA architectures. Focus on addressing constraints such as memory management, latency, and system dependencies.
Demonstrate Linux and Automation Proficiency
Prepare a repository of Bash and Python scripts that automate the deployment or troubleshooting of a local AI environment. Include examples of managing system dependencies, configuring networking components like port forwarding, and handling secure traffic.
Formulate Strategic Questions for Stakeholders
Prepare questions that demonstrate your ability to bridge the gap between technical AI capabilities and business needs. Focus on how to identify high-impact use cases for enterprise clients, manage executive expectations, and scale internal AI solutions.
Review Networking and Security Fundamentals
Brush up on your knowledge of TCP/IP, DNS, and webhooks in the context of AI deployments. Be ready to discuss how to secure local AI environments when integrating them into broader enterprise network architectures.

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…