New opportunity

Solutions Engineer - Artificial Intelligence (AI) Practitioner

Cisco · Singapore

About this role

Cisco seeks a Solutions Engineer - Artificial Intelligence (AI) Practitioner for its sales team in Singapore. The role is hybrid and full-time, with a stated schedule of 40 hours per week. Within the Cloud and AI Infrastructure organization, the engineer will help customers adopt AI solutions across industries, connecting technical designs to business outcomes.

Responsibilities

Consult with prospective customers to understand their needs and develop tailored, deployable AI solutions.
Collaborate with customers, partners, sales teams, product specialists, and product engineering teams on challenges involving AI, data centers, networking, and cloud technologies.
Explain model architectures, infrastructure requirements, and business outcomes to audiences ranging from data scientists and platform teams to executives. Lead customer workshops, architecture discussions, hands-on demonstrations, and technical evaluations.

Minimum qualifications:

10+ years of experience in technology consulting and pre-sales, preferably in software, data, analytics, infrastructure, AI, or related fields.
Strong understanding of programming and scripting, particularly Python; machine learning, deep learning, generative AI, and modern AI application architectures.
Experience designing and documenting enterprise technology solutions using established architecture and deployment methodologies, developing practical AI solutions, and working with AI infrastructure such as GPU systems, networking, storage, cloud services, or container platforms.
Excellent presentation and communication skills for technical and non-technical audiences.

Preferred qualifications

A bachelor's degree in Computer Science, Computer Engineering, Data Science, or a related field; an advanced degree is a plus.
Experience with GPU accelerators, InfiniBand or RoCE networking, scale-out storage, and model-serving or inference technologies such as vLLM, NVIDIA NIM, Triton Inference Server, or TensorRT-LLM.
In-depth understanding of transformer-based language models and multimodal architectures; experience with PyTorch, TensorFlow, Jupyter, and NVIDIA AI Enterprise; and familiarity with Kubernetes, OpenShift, containerization, MLOps, and production AI operations.

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

Skills for this role

PythonSolution architectureTechnology consultingPre-salesMachine learningDeep learningGenerative AIAI application architectureAI infrastructureGPU systemsNetworkingStorageCloud servicesContainer platformsCustomer workshopsTechnical demonstrationsTechnical evaluationsPresentationCommunicationInfiniBandRoCEScale-out storagevLLMNVIDIA NIMTriton Inference ServerTensorRT-LLMTransformer-based language modelsMultimodal architecturesPyTorchTensorFlow

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

Translate customer needs into deployable AI solutions
The role consults prospective customers, identifies their needs, and develops tailored solutions tied to business outcomes.
Design and document enterprise AI architecture
The minimum qualifications call for established architecture and deployment methodologies, practical AI application design, and experience with GPU systems, networking, storage, cloud services, or container platforms.
Lead hands-on customer evaluations
The engineer is expected to lead workshops, architecture discussions, demonstrations, and technical evaluations, and to demonstrate AI solutions that address real business problems.
Communicate across technical and executive audiences
The posting emphasizes explaining model architectures to data scientists, infrastructure requirements to platform teams, and business outcomes to executives.
Coordinate across customer and internal teams
The role works with customers, partners, sales teams, product specialists, and product engineering to turn customer problems into deployable solutions.

Likely challenges to think through

Balancing ambition with deployability
Preparation hypothesis: A compelling AI design may be hard to deploy within infrastructure constraints. Prepare an example of narrowing a use case, documenting architecture assumptions, and explaining GPU, network, storage, and deployment trade-offs without treating every preferred technology as mandatory.
Making a demonstration credible
Preparation hypothesis: A hands-on demo may show promise without proving suitability for production. Prepare to distinguish what a prototype demonstrates from what still needs technical evaluation, and to connect both to the customer's business problem.
Aligning different decision-makers
Preparation hypothesis: Data scientists, platform teams, and executives may value different outcomes. Prepare one architecture explanation at three levels of detail, showing how technical choices affect operational feasibility and business value.

Practice your answers

Technical practice

How would you turn a prospective customer’s AI goal into a deployable solution?

What to coverThe role involves understanding customer needs and tailoring AI solutions to business outcomes. Suggested preparation: explain how you would clarify the use case, data and infrastructure constraints, then document an architecture and criteria for success.

How would you choose between conventional machine learning and a generative AI approach?

What to coverThe posting calls for an understanding of machine learning, deep learning, generative AI and modern AI application architectures. Suggested preparation: compare approaches against the task, available data, evaluation needs and operational complexity rather than assuming generative AI is always appropriate.

What architectural considerations change when moving from a transformer-based language model to a multimodal application?

What to coverTransformer-based language models and multimodal architectures are listed as preferred knowledge. Suggested preparation: discuss input processing, model and serving requirements, and how you would evaluate outputs for the intended use case.

How would you build a Python-based demonstration of a practical AI solution?

What to coverPython proficiency and the ability to develop and demonstrate practical AI solutions are stated qualifications. Suggested preparation: outline a small, reproducible workflow that connects a business problem to an AI application, and explain what the demonstration would and would not prove.

How would you size and connect GPU, networking and storage resources for an AI workload?

What to coverThe role covers AI infrastructure; GPU accelerators, InfiniBand or RoCE, and scale-out storage appear as preferred experience. Suggested preparation: discuss workload characteristics, data movement and bottlenecks, making clear which design choices depend on measurements.

How would you compare options for serving a model in an enterprise solution?

What to coverThe posting names vLLM, NVIDIA NIM, Triton Inference Server and TensorRT-LLM as examples of preferred inference experience. Suggested preparation: compare relevant options using model compatibility, latency, throughput, resource use and operational fit without claiming one is universally best.

What would you consider when moving an AI application from a demonstration into production operations?

What to coverContainerization, Kubernetes, OpenShift, MLOps and production AI operations are listed as preferred familiarity. Suggested preparation: describe deployment, monitoring, versioning and recovery considerations, distinguishing a working demonstration from an operable service.

How would you document an AI solution architecture for both platform teams and decision-makers?

What to coverThe posting calls for designing and documenting enterprise solutions and explaining infrastructure requirements and business outcomes to different audiences. Suggested preparation: describe the diagrams, assumptions, dependencies and trade-offs you would present at each level of detail.

How would you structure a customer technical evaluation of a proposed AI solution?

What to coverLeading technical evaluations and connecting solutions to business outcomes are stated duties. Suggested preparation: propose a scoped test, agreed success measures and a way to report limitations alongside technical results.

How would you diagnose an AI demonstration that misses its performance target?

What to coverHands-on demonstrations and AI infrastructure are part of the advertised work. Suggested preparation: explain how you would isolate model, serving, GPU, network and storage factors, measure before changing the design, and communicate the implications to the customer.

Behavioural practice

Tell me about a time you led a workshop to clarify an ambiguous customer problem.

What to coverLeading customer workshops and understanding customer needs are stated duties. Suggested preparation: use a real situation, explain how you elicited and reconciled requirements, and describe the resulting decision or outcome.

Describe a time you explained the same technical proposal to specialists and executives.

What to coverThe role calls for communication with data scientists, platform teams and executives. Suggested preparation: give a real situation, the changes you made to your explanation for each audience, and the result.

Tell me about a time you coordinated several teams to resolve a solution-design disagreement.

What to coverThe posting describes collaboration with customers, partners, sales teams, product specialists and engineering teams. Suggested preparation: identify a real disagreement, your actions to clarify trade-offs and ownership, and the outcome.

Describe a time a hands-on demonstration did not go as planned.

What to coverHands-on demonstrations are part of the advertised role. Suggested preparation: recount the situation, how you diagnosed and communicated the issue, and what resulted; avoid presenting a hypothetical recovery as personal experience.

Tell me about a technical evaluation that challenged your initial recommendation.

What to coverThe role includes technical evaluations and tailoring solutions to customer needs. Suggested preparation: use a real example to explain the evidence you received, how you adjusted or defended the recommendation, and the final outcome.

Sharpen your knowledge

Topics to prepare

ML, deep learning and generative AI architectures
Minimum qualification: understand these approaches and modern AI application architectures. Revision exercise: sketch an AI application for a hypothetical customer need, explaining the model’s role, infrastructure dependencies and business outcome.
Enterprise solution and deployment architecture
Minimum qualification: design and document enterprise solutions using established architecture and deployment methodologies. Revision exercise: review an architecture diagram and identify its components, deployment steps, dependencies and assumptions.
AI infrastructure trade-offs
Minimum qualification: experience with AI infrastructure including GPU systems, networking, storage, cloud services or container platforms. Revision exercise: explain how changes in workload size or latency needs might affect infrastructure choices. InfiniBand, RoCE and scale-out storage are preferred, not minimum, qualifications.
Transformers, multimodal models and inference
Preferred qualifications: understand transformer-based language models and multimodal architectures, and have experience with model-serving technologies. Revision exercise: trace a request from input through inference to response, then explain where serving capacity and latency matter.
Production AI operations
Preferred qualification: familiarity with Kubernetes, OpenShift, containerization, MLOps and production AI operations. Revision exercise: outline what a platform team would need to deploy, monitor and update a model-serving application; treat this as preparation, not a stated assignment.

Skills to revise

Python AI demonstration
Minimum qualifications call for strong Python understanding and the ability to develop and demonstrate practical AI solutions. Practice walking through a small Python-based AI demonstration, explaining its inputs, outputs, limitations and business use without implying a particular demo is required.
Architecture documentation
The posting requires experience designing and documenting enterprise solutions. Practice producing a one-page diagram and deployment outline for a hypothetical customer AI use case, labeling assumptions and decisions for review.
Customer discovery workshop
The role involves understanding customer needs and leading workshops. Rehearse questions that elicit the business problem, intended users, existing infrastructure and success criteria; then summarize the answers as a deployable-solution brief.
Technical evaluation facilitation
The role includes leading technical evaluations. Practice proposing evaluation criteria for a hypothetical AI solution, then explain how you would discuss results and unresolved trade-offs with a customer. The criteria are a preparation suggestion, not a specified Cisco process.
Audience-specific presentation
The posting calls for communication with data scientists, platform teams and executives. Practice presenting the same proposed solution three ways: model architecture, infrastructure requirements and business outcome, respectively.

Your preparation checklist

Map the role to evidence
Prepare a one-page evidence map for the posting’s 10+ years of consulting and pre-sales experience, enterprise solution design, Python, practical AI work, and customer-facing presentations. Use specific examples where available; leave gaps as gaps.
Build a customer project walkthrough
Select a relevant past AI engagement, if available, and prepare a concise walkthrough of customer needs, your role, architecture and deployment decisions, technical evaluation, collaborators, and business outcome. The posting emphasizes turning customer problems into deployable solutions.
Practice a tailored architecture exercise
For a hypothetical customer AI use case, draw a deployable design covering the application and model, GPU capacity, networking, storage, cloud or containers, and deployment trade-offs. Label assumptions and explain how you would validate them with the customer.
Prepare a practical AI demonstration
Outline or rehearse a short Python-based demo that addresses a business problem. Be ready to explain the model or generative-AI architecture, infrastructure needs, limitations, and evaluation approach; the role calls for hands-on demonstrations and technical evaluations.
Rehearse a customer workshop
Draft discovery questions and a workshop agenda, then practice explaining the same proposed solution to data scientists, platform teams, and executives. Prepare distinct explanations of model choices, infrastructure requirements, and business outcomes.
Review preferred technologies selectively
For technologies you have used, prepare precise examples involving GPU accelerators, InfiniBand or RoCE, scale-out storage, model serving, or production AI operations. These are preferred qualifications, not a requirement to claim experience with every named tool.
Bring questions for Cisco
Write questions about typical customer AI use cases, how the Solutions Engineer works with sales, partners, product specialists, and engineering, and what makes a technical evaluation successful. The posting describes those collaborations but does not specify a standard evaluation process.

Availability last checked: 4 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 Singapore. Matched to this role, not your profile.

Explore more jobs

Finding similar opportunities…