AI Soution Architect

Job Description:

Responsibilities

• AI/ML & GenAI Architecture
• Design end-to-end AI architectures for use cases such as semantic search, recommendation systems, conversational AI, forecasting, and intelligent automation
• Architect and implement RAG-based systems using LLMs, embedding models, vector databases, and orchestration frameworks
• Define design patterns for multi-agent systems, tool-augmented LLMs, and agentic workflows
• Model Development & Optimization
• Develop, fine-tune, and evaluate ML/DL models for classification, regression, ranking, and recommendation use cases
• Optimize LLM pipelines for latency, cost, and accuracy through prompt engineering, caching, and retrieval strategies
• Implement evaluation frameworks for LLM outputs (hallucination detection, grounding, relevance scoring)
• Data Engineering & Pipelines
• Build scalable data pipelines using PySpark, Spark, or distributed systems
• Design feature engineering and data preprocessing workflows for structured and unstructured data
• Work with real-time and batch data processing systems (Kafka, streaming frameworks)
• Search & NLP Systems
• Design hybrid search systems combining lexical (BM25, Elasticsearch/OpenSearch) and semantic (embeddings, ANN) approaches
• Build NLP pipelines for entity extraction, summarization, classification, and Q&A systems
• Deployment & MLOps
• Build and deploy APIs for model serving using FastAPI, Flask, or similar frameworks
• Implement CI/CD pipelines for ML systems, including versioning, monitoring, and rollback strategies
• Deploy models using Docker, Kubernetes, and cloud-native services
• Monitor model performance, drift, and system health in production
• Cloud & Platform Engineering
• Architect AI systems on AWS, GCP, or Azure using services such as SageMaker, Bedrock, and Vertex AI
• Design scalable, fault-tolerant, and cost-efficient cloud architectures
• Leverage managed vector databases and data lake architectures
• Collaboration & Leadership
• Translate business problems into scalable AI solutions and technical designs
• Collaborate with product, engineering, and data teams to deliver production-ready systems
• Mentor engineers and establish best practices for AI/ML engineering and architecture

Qualification & Experience  

• Core Programming & Engineering
• Strong proficiency in Python (advanced OOP, async programming, performance optimization)
• Experience building production-grade backend systems and APIs
• Machine Learning & Deep Learning
• Hands-on experience with PyTorch, TensorFlow, or Scikit-learn
• Strong understanding of supervised/unsupervised learning, evaluation metrics, feature engineering, and model tuning
• Generative AI & LLMs
• Experience with OpenAI, Anthropic, or open-source LLMs (LLaMA, Mistral, etc.)
• Strong expertise in prompt engineering, RAG pipelines, and embedding models
• Experience with vector databases (Pinecone, Weaviate, FAISS, Milvus)
• Search & Retrieval
• Experience with Elasticsearch/OpenSearch
• Understanding of hybrid retrieval and ranking strategies
• Data & Distributed Systems
• Strong experience with NumPy, Pandas, and PySpark
• Experience handling large-scale datasets and distributed processing
• MLOps & Deployment
• Experience with FastAPI/Flask for model serving
• Hands-on experience with Docker, Kubernetes, and CI/CD pipelines (GitHub Actions, Jenkins, etc.)
• Familiarity with model monitoring, logging, and observability
• Cloud Platforms
• Hands-on experience with AWS, GCP, or Azure
• Understanding of cloud-native AI/ML services and infrastructure
• Qualifications
• Bachelor’s or master’s degree in computer science, AI, or related field.
• Good to Have:
• Experience with LangChain, LangGraph, Dify, LlamaIndex, or similar frameworks
• Experience designing multi-agent or autonomous AI systems
• Knowledge of graph databases (Neo4j) and knowledge graphs
• Experience with streaming systems (Kafka, Flink)
• Exposure to Computer Vision, Time-Series Forecasting, or Reinforcement Learning
• Experience in cost optimization and scaling GenAI systems in production