AI Architect

Skill Required: AIML, RAG, GENAI, LangChain, Rest API, GraphQL,

hands-on experience in building solutions using AWS Bedrock and ecosystem.

Key Responsibilities

Design multilayered AI solutions balancing compute efficiency contextual fidelity and algorithmic adaptability retrieval reasoning planning tooluse

Develop advanced RAG pipelines leveraging vector databases eg ChromaDB Milvus FAISS and embedding strategies for contextual accuracy

Integrate AI capabilities with enterprise systems via REST and GraphQL APIs ensuring secure and scalable interoperability

Establish best practices for algorithm selection and layering combining neural models symbolic reasoning and toolbased agents for optimal performance

Collaborate with crossfunctional teams to embed AI agents into business workflows and align with compliance and governance standards

Implement structured output validation and schema enforcement using Pydantic FastAPI and JSON Schema for robust data integrity

Optimize compute resources and latency tradeoffs across cloud hybrid and edge environments for highperformance AI workloads

Define observability baselines telemetry tracing evaluation metrics and rollout strategies for safe iterative deployments

Required Skills

Programming Architecture Python 310 Async design modulardistributed architecture microservices

LangChain Ecosystem LangChain LangGraph prompt templates agent orchestration patterns

AIML Frameworks OpenAI API HuggingFace Transformers TensorFlow PyTorch experience with finetuning and inference optimization

Data Context Management SQLAlchemy PostgreSQL JSON Schema Mapping feature engineering and contextual pipelines

Vector Databases Semantic Search ChromaDB Milvus FAISS embedding optimization and similarity search strategies

Algorithmic Design Strong understanding of algorithm layering retrieval reasoning planning hybrid AI approaches and computeaware model selection

API Integration Security REST GraphQL OAuth enterprisegrade security practices

DevOps CICD Git Docker Azure DevOps or equivalent FastAPI Uvicorn containerized deployments and automated pipelines

Agentic Capabilities Reasoning adaptation tool calling MCPbased solutions ReAct agents Supervisormultiagent coordination

Performance Optimization Distributed compute strategies GPUTPU utilization quantizationpruningdistillation caching batching