New opportunity

Principal Engineer, AI Platform

Accrete · Mumbai, India

About this role

Accrete seeks a hands-on Principal Engineer for its AI Platform team in Andheri East, Mumbai. The role leads the architecture, development and scaling of the platform behind its AI products and agentic systems, working across frontend, backend and infrastructure. The posting specifies hybrid work, with remote flexibility under company guidelines.

Responsibilities

Architect secure, reliable and maintainable platform components for complex AI agent workloads. Set technical direction, resolve architectural bottlenecks and weigh scalability, performance, reliability, cost and time-to-market.
Build end-to-end features, APIs, services, core libraries, developer tools and interfaces for AI models, agents and data. Develop infrastructure to run and orchestrate agents securely in production, and integrate AI/ML models and data pipelines.
Improve availability, fault tolerance, latency, resource utilization, observability and developer experience. Establish practices for design, code quality, testing, security and operations.
Lead design and code reviews, mentor engineers, contribute to the platform roadmap and own initiatives from architecture through deployment and iteration. Collaborate with engineering, ML, data science and product teams.

Required qualifications

At least 10 years of software engineering experience, including designing scalable production systems and leading complex initiatives. The qualifications section accepts a PhD in computer science, AI, machine learning or a related field, or a master's degree in a related technical field from a Tier I institution; the posting also separately labels the role 'Ph.D Required.'
Strength in software architecture, distributed systems and technical leadership; frontend experience with React and TypeScript or similar technologies; backend experience with Python or Go; and knowledge of REST, GraphQL, databases, AWS, GCP or Azure, Docker, Kubernetes, CI/CD, observability and production operations. Familiarity with AI/ML frameworks, model serving or AI application architecture is sought.

Preferred experience includes AI agent platforms and orchestration, LLMs, generative AI, PyTorch, TensorFlow, real-time or streaming systems, Apache Spark, secure execution environments, developer platforms, and security, compliance, privacy or government-focused work. Benefits listed include ₹10 lakh family medical coverage, flexible PTO, learning opportunities, catered food and team events; compensation is described as competitive without an amount.

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

Skills for this role

AI platform architectureAI agentsAgent orchestrationSoftware architectureSystem designDistributed systemsFull-stack developmentReactTypeScriptPythonGoRESTGraphQLAPI designRelational databasesNoSQLAWSGCPAzureDockerKubernetesCloud-native architectureCI/CDDevOpsObservabilitySecurityModel servingData pipelinesTechnical leadershipMentoring

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

Set AI platform architecture
Lead core platform design and technical direction across frontend, backend and infrastructure. The posting emphasizes resolving bottlenecks and weighing scalability, performance, reliability, cost and time-to-market.
Productionize agent workloads
Build infrastructure to run and orchestrate agents securely, integrate AI/ML models and data pipelines, and keep platform components modular as AI capabilities evolve.
Deliver full-stack platform capabilities
Own end-to-end features, APIs, services, core libraries and developer-facing interfaces for working with models, agents and data.
Raise production and engineering standards
Improve availability, fault tolerance, latency, resource utilization, observability and developer experience while establishing practices for design, testing, security and operations.
Lead through delivery
Own initiatives from architecture through deployment and iteration; lead reviews, mentor engineers, shape the roadmap and collaborate with engineering, ML, data science and product teams.

Likely challenges to think through

Secure agent execution without blocking progress
Likely preparation hypothesis: agents acting on high-stakes work create tension between useful capabilities and controlled production execution. Prepare to explain how you would define boundaries, test failure modes and make security trade-offs explicit; the posting does not prescribe an implementation.
Balance operational goals against delivery speed
Likely preparation hypothesis: improving latency, availability and resource use simultaneously can conflict with cost and time-to-market. Prepare a concrete approach to measuring bottlenecks, prioritizing changes and explaining trade-offs.
Stay hands-on while directing multiple teams
Likely preparation hypothesis: ownership across the stack can compete with time needed for reviews, mentoring and cross-functional alignment. Prepare to describe how you would delegate decisions, preserve architectural coherence and follow an initiative through deployment.

Practice your answers

Technical practice

How would you architect a platform for complex AI agent workloads?

What to coverThe role calls for scalable, secure and maintainable systems, with decisions balancing performance, reliability, cost and time-to-market. Suggested preparation: explain component boundaries, likely bottlenecks and how you would evaluate those trade-offs as workloads grow.

How would you run and orchestrate agents securely in production?

What to coverThe posting explicitly includes secure agent orchestration; secure execution of untrusted workloads is listed as a nice-to-have. Suggested preparation: discuss isolation, permissions, failure handling and the operational cost of stronger safeguards without assuming a particular implementation.

How would you make organizational context usable and auditable for agents?

What to coverAccrete describes a dynamic context graph that reconciles data across silos and provides an auditable ground truth. Suggested preparation: consider provenance, access control, data freshness and how an agent-facing interface could expose uncertainty or conflicting records.

How would you design an end-to-end feature for interacting with models and agents?

What to coverThe role spans frontend and backend development, including developer-facing tools and interfaces; React, TypeScript, Python and Go are named technologies. Suggested preparation: walk through the interface, service boundaries, error handling and how you would keep the feature usable and maintainable.

When would you choose REST or GraphQL for a platform API?

What to coverThe posting asks for understanding of REST, GraphQL and API/service architectures. Suggested preparation: compare client needs, schema evolution, authorization, performance and operational complexity for an agent or developer-facing API.

How would you integrate data pipelines with a production AI platform?

What to coverIntegrating models and data pipelines is a stated duty, and relational or NoSQL database experience is requested. Suggested preparation: explain how you would choose storage and data interfaces, validate inputs, handle updates and detect failures without assuming a specified pipeline technology.

How would you integrate model serving into a modular platform?

What to coverThe role seeks familiarity with model-serving infrastructure or AI application architectures and calls for components that can evolve with AI technology. Suggested preparation: discuss interface boundaries, version changes, latency and reliability trade-offs between platform services and model-serving components.

How would you deploy and scale platform services in a cloud-native environment?

What to coverThe posting names AWS, GCP or Azure alongside Docker, Kubernetes and cloud-native architectures. Suggested preparation: explain deployment boundaries, capacity planning, availability and fault tolerance, while distinguishing general design choices from any cloud configuration the posting does not specify.

How would you investigate rising agent-workload latency and resource use?

What to coverImproving latency, resource utilization, reliability and observability is part of the role. Suggested preparation: describe the signals you would inspect, how you would isolate a bottleneck across services and what evidence would justify a performance or cost trade-off.

How would you establish release and quality practices for a changing AI platform?

What to coverThe role includes setting practices for design, testing, security and operations, and asks for CI/CD and production-operations knowledge. Suggested preparation: discuss review and test boundaries, deployment checks, rollback decisions and how to improve delivery speed without weakening reliability.

Behavioural practice

Tell me about an initiative you owned from architecture through production.

What to coverThe posting emphasizes end-to-end ownership and accountability for technical outcomes. Suggested preparation: use a real situation/action/result example covering your decisions, deployment and subsequent iteration, including what changed as a result.

Describe a time you resolved an ambiguous architectural bottleneck.

What to coverThe role calls for independent judgment about technical risks and trade-offs. Suggested preparation: give a real situation/action/result account of how you gathered evidence, compared options, gained agreement and assessed the outcome.

Tell me about mentoring an engineer or raising a team's engineering standards.

What to coverMentorship, design reviews and code reviews are stated responsibilities. Suggested preparation: use a real situation/action/result example that shows the guidance you gave, how you handled feedback and an observable change in practice.

Describe a cross-functional disagreement about a platform requirement.

What to coverThe role collaborates with product, ML, data science and engineering teams to turn requirements into technical solutions. Suggested preparation: explain a real situation/action/result example showing how you listened, communicated constraints, reached a decision and measured its effect.

Tell me about evaluating an emerging AI technology without losing focus on delivery.

What to coverThe posting includes evaluating emerging AI/ML technologies for the platform roadmap. Suggested preparation: use a real situation/action/result example covering your evaluation criteria, communication of uncertainty and decision to adopt, defer or reject an approach.

Sharpen your knowledge

Topics to prepare

Distributed platform architecture
Posting: The role designs scalable, reliable AI Platform components and resolves bottlenecks. Revision exercise: Sketch a hypothetical multi-service agent workload, identify failure points, and explain trade-offs among latency, availability, cost, and time-to-market.
Secure AI agent orchestration
Posting: The role builds infrastructure to run and orchestrate agents securely in production; secure execution environments are nice to have. Revision exercise: Describe trust boundaries and failure handling for a hypothetical agent invoking a platform service. Treat the design as interview practice, not Accrete's specified architecture.
API, data, and model integration
Posting: The role builds APIs and services for agents and integrates AI/ML models and data pipelines; REST, GraphQL, and databases appear in technical expertise. Revision exercise: Compare REST and GraphQL interfaces for a hypothetical agent-and-data feature, including data access and model-serving boundaries.
Cloud-native reliability and operations
Posting: AWS, GCP or Azure; Docker; Kubernetes; CI/CD; observability; and production operations are listed. Revision exercise: Trace how a hypothetical platform service is deployed, monitored, and recovered after a dependency failure; explain which signals reveal latency or resource-utilization problems.
Principal-level technical judgment
Posting: The engineer leads reviews, mentors engineers, influences direction across teams, and owns initiatives through deployment and iteration. Revision exercise: Rehearse an architecture review in which two viable designs conflict; state a recommendation, its risks, how you would gather input, and how you would revisit the decision.

Skills to revise

Backend service implementation
Posting: Strong backend experience with Python, Go, or similar languages and experience with API/service architectures are sought. Practice suggestion: Implement a small service endpoint in Python or Go for a hypothetical agent task; add input validation, error handling, and tests.
Frontend-to-backend feature delivery
Posting: The role builds end-to-end features and developer-facing interfaces; React and TypeScript or similar technologies are listed. Practice suggestion: Build a small interface that submits a hypothetical agent task to an API and displays status and errors.
Container deployment and diagnosis
Posting: Docker, Kubernetes, cloud-native architectures, CI/CD, and observability are named. Practice suggestion: Containerize the practice service, outline a Kubernetes deployment and CI/CD checks, then walk through diagnosing a failing release using logs and metrics.
Data and model integration
Posting: Integrating AI/ML models and data pipelines is a responsibility; relational and/or NoSQL databases and familiarity with model-serving infrastructure are listed. Practice suggestion: Connect a sample service to a database and a stubbed model-serving endpoint; test unavailable or slow dependencies. The stub is a preparation aid, not an employer-specified tool.
Design and code review facilitation
Posting: The role leads technical design and code reviews and mentors engineers. Practice suggestion: Review a hypothetical platform change, write prioritized feedback on security, reliability, tests, and maintainability, then explain the trade-offs to both engineering and product stakeholders.

Your preparation checklist

Map the role to evidence
Prepare a one-page reference to projects or experience you can discuss across scalable production systems, distributed architecture, full-stack development and technical leadership. The posting specifies 10+ years and lists both a PhD requirement and a PhD-or-Tier-I-master’s qualification; note that discrepancy for clarification.
Build a production-project walkthrough
Choose one initiative you owned from architecture through deployment and iteration. Prepare a diagram, your decisions, alternatives considered, measurable outcomes and what you changed after launch.
Practice an agent-platform design exercise
Sketch a hypothetical platform for securely running and orchestrating agents. Explain APIs, model and data integration, isolation, failure handling and trade-offs among scalability, latency, reliability, cost and delivery time. This is a preparation exercise, not a claimed Accrete design.
Trace an end-to-end feature
Prepare a feature walkthrough from a developer-facing interface to an API, service and data store. Be ready to explain choices involving React or TypeScript, Python or Go, and REST or GraphQL where they match your experience.
Show how you would operate the platform
Create a brief operations plan covering deployment, testing, CI/CD, observability, incident diagnosis and improvements to availability or resource use. Use a past example and its results if available.
Prepare leadership examples
Write concise examples of a design review, mentoring decision and cross-team trade-off involving engineering, ML, data or product colleagues. Identify your contribution and the resulting technical outcome.
Draft questions for Accrete
Ask how the AI Platform interacts with the described Knowledge Engine and expert agents, which reliability or developer-experience problems have priority, and how the education wording is interpreted for this role.

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…