Overview

Location: India - Delhi

Responsibilities

• Own the technical architecture of AI-enabled enterprise products from discovery to production.
• Define application boundaries, service contracts, data models, integration patterns, deployment topology, and non-functional requirements.
• Select appropriate technologies for LLM orchestration, data processing, vector search, workflow automation, and enterprise integration.
• Produce architecture decision records, solution blueprints, sequence diagrams, API specifications, threat models, capacity plans, and operational runbooks.
• Establish reusable reference architectures for RAG, AI agents, human-in-the-loop review, model gateways, and workflow automation.
• Define standards for API design, schema validation, error handling, observability, testing, secure coding, versioning, and release management.
• Review designs and code produced by other engineers.
• Lead proof-of-concepts and convert successful prototypes into production-grade implementations.
• Balance functional requirements, performance, reliability, security, cost, maintainability, and delivery timelines.
• Develop high-performance Python services using FastAPI, Django, Flask, or equivalent frameworks.
• Build asynchronous processing pipelines using Kafka, Redis Streams, AWS SQS/SNS, Azure Service Bus, or equivalent technologies.
• Develop frontend applications using React, Next.js, JavaScript, HTML5, CSS3, and/or SAPUI5.
• Build enterprise dashboards, AI workbenches, workflow consoles, document-review interfaces, search applications, approval journeys, and administrative tools.
• Implement streaming AI responses, structured result views, confidence indicators, citations, source references, audit trails, and human feedback mechanisms.
• Design and implement analytical, document, metadata, and vector data stores.
• Build reusable shared libraries, SDKs, frontend components, API clients, authentication modules, and platform services.
• Develop secure integrations with internal systems, third-party services, data platforms, and enterprise identity providers.
• Design and productionize RAG systems using document ingestion, chunking, metadata enrichment, embeddings, reranking, retrieval filters, and response generation.
• Implement AI agents and tool-using workflows with controlled planning, execution, validation, retries, approvals, and escalation.
• Integrate LLMs from approved providers and model platforms, including enterprise model gateways and SAP Generative AI Hub where applicable.
• Implement prompt versioning, model versioning, evaluation datasets, golden test cases, regression suites, and quality gates.
• Establish controls for hallucination, prompt injection, data leakage, unsafe tool use, unauthorized access, and excessive model autonomy.
• Design model fallback, timeout, retry, caching, token-budget, and cost-control strategies.
• Implement observability for prompts, responses, model metadata, tool calls, latency, token usage, retrieval quality, user feedback, and failure modes.
• Collaborate with data scientists and platform engineers on model training, fine-tuning, inference optimization, deployment, and monitoring.
• Build AI features that support structured extraction, classification, summarization, recommendations, forecasting, conversational search, and business-process automation.
• Lead the design and implementation of SAP BTP extension and side-by-side applications.
• Design integrations between SAP BTP, SAP S/4HANA, SAP ECC, SAP SuccessFactors, SAP Ariba, SAP Analytics Cloud, and non-SAP systems.
• Use OData, REST, RFC, events, SAP Gateway, and SAP Integration Suite as appropriate.
• Work with SAP BTP services such as:

Qualifications

• Location: India - Delhi
• The candidate must have experience taking products from requirements and architecture through development, testing, deployment, production operations, enhancement, and stakeholder adoption with at least 3-4 complete, end-to-end production implementations.
• Bachelors or masters degree in computer science, Software Engineering, Information Technology, Data Science, or a related discipline.
• Advanced Python and strong software design ability.
• FastAPI, Django, Flask, Pydantic, SQLAlchemy, and asynchronous programming.
• TypeScript/Node.js for CAP services and enterprise integrations.
• REST, GraphQL where appropriate, OpenAPI, WebSockets, event-driven systems, and messaging.
• PostgreSQL, SAP HANA Cloud, MongoDB, Redis, Elasticsearch/OpenSearch, and vector databases.
• Microservices, modular monoliths, distributed systems, caching, idempotency, retries, circuit breakers, and resiliency patterns.
• Advanced JavaScript, React, Next.js, component architecture, state management, frontend security, and testing.
• SAPUI5 and Fiori development or the ability to lead SAP UI implementation.
• Accessibility, responsive design, performance optimization, design systems, and user-centered AI experiences.
• LLM application architecture and production integration.
• RAG, embeddings, reranking, vector databases, semantic search, hybrid search, and document processing.
• Agentic workflows, function calling, tool execution, workflow graphs, and human approval patterns.
• Prompt engineering, structured generation, JSON schema validation, guardrails, and model routing.
• Model evaluation, benchmark creation, regression testing, red teaming, safety testing, and feedback analysis.
• Fine-tuning or parameter-efficient adaptation of open-source models is preferred.
• Understanding of GPU inference, model serving, batching, quantization, latency, throughput, and cost optimization is advantageous.
• Strong experience in SAP BTP application deployment using CDS, SAP HANA Cloud, Cloud Foundry, and/or Kyma.
• SAP AI Core, AI Launchpad, Generative AI Hub, Document AI, and SAP Business AI.
• SAP S/4HANA integration using OData, REST, RFC, events, and Integration Suite.
• SAPUI5/Fiori, SAP Gateway, OAuth 2.0, OpenID Connect, SAML, XSUAA, role collections, technical users, and enterprise.
• Understanding of SAP clean-core, side-by-side extensibility, tenant isolation, and lifecycle management.
• SAP BTP or SAP AI certification is preferred.
• AWS, Azure, Google Cloud, and/or SAP BTP.
• Docker, Kubernetes, Cloud Foundry, and infrastructure-as-code.
• CI/CD with GitHub Actions, GitLab, Azure DevOps, or equivalent.
• OpenTelemetry, logging, metrics, distributed tracing, dashboards, and alerting.
• Secrets management, vulnerability scanning, dependency management, image scanning, and software supply-chain security.