About the role
- Generative AI – production-grade GenAI solution design and deployment
- Agentic AI / Multi-Agent Systems – agent orchestration, tool-using agents, memory-enabled systems
- Advanced RAG Architecture – multi-stage retrieval, re-ranking, multi-hop retrieval and reasoning
- Python (core), FastAPI, React
- Vector Databases & Embedding Models – hybrid search architectures
- LLM Integration – LLM APIs, multi-model AI architectures
- AI Platform Engineering – model serving, feature stores, GPU/infra readiness
- Production AI Deployment – observability, logging, tracing, reliability engineering, graceful degradation, circuit breakers
- AI Evaluation Frameworks – A/B testing, benchmarking, telemetry-based optimization
- Prompt Engineering – templates, versioning, testing methodologies
- Observability & Monitoring – real-time dashboards, automated alerting, incident response
- Enterprise/Cloud-Native Architecture – distributed systems at scale
- Cloud AI Platforms – GCP/Azure
- With loops & graphs hands on
- Architect end-to-end AI systems including advanced RAG pipelines, multi-agent orchestration frameworks, and multi-model AI integrations built for modularity, scalability, and operational excellence.
- Define enterprise standards for prompt engineering (templates, versioning, testing, evaluation) and performance optimization (model selection, caching, resource utilization, cost).
- Lead deployment of AI solutions into production with comprehensive observability, reliability engineering, monitoring dashboards, automated alerting, and incident response — meeting stringent SLOs.
- Design scalable data ingestion frameworks for structured, unstructured, and real-time streaming data, along with vector database architectures, hybrid search, preprocessing pipelines, and data quality/governance frameworks.
- Establish quantitative AI evaluation frameworks (A/B testing, benchmarking, user feedback, telemetry) and drive continuous improvement across prompts, retrieval strategies, agent workflows, and model configurations.
- Partner with platform and infrastructure teams on AI workload readiness (GPU infra, model serving, feature stores, storage, networking) and define enterprise AI platform requirements.
- Ensure AI solutions adhere to enterprise governance and compliance; apply Responsible AI principles — fairness, transparency, accountability, and bias mitigation.