Senior AI Engineer / Agentic AI Architect

About the role

Requirements

• Requirements
• Required Qualifications & Skills
• Experience
• - 8–12+ years of experience in Software Engineering.
• - 5+ years of hands-on experience in Artificial Intelligence, Machine Learning, and Generative AI.
• AI / Machine Learning
• - Strong understanding of:
• - Machine Learning
• - Deep Learning
• - NLP
• - Transformer architectures
• - Large Language Models (LLMs)
• - Embedding models
• Software Engineering
• - Expert-level Python programming.
• - Strong software engineering fundamentals.
• - Experience building production-grade backend systems.
• - RESTful API and microservices development.
• - Async programming and scalable architectures.
• - Experience with FastAPI, Flask, or similar frameworks.
• Agentic AI
• - Hands-on experience designing and implementing Agentic AI solutions using one or more of:
• - LangGraph
• - CrewAI
• - OpenAI Agents SDK
• - AutoGen
• - Semantic Kernel
• - LlamaIndex Workflows
• - Experience in:
• - Multi-agent orchestration
• - Planning agents
• - Tool calling
• - Human-in-the-loop workflows
• - Memory management
• - State management
• - Agent collaboration patterns
• Retrieval-Augmented Generation (RAG)
• - Strong experience building enterprise RAG platforms.
• - Embedding models, such as:
• - OpenAI
• - Voyage AI
• - BGE
• - E5
• - Instructor
• - Cohere
• - Vector databases, such as:
• - Pinecone
• - Qdrant
• - Milvus
• - Weaviate
• - ChromaDB
• - Graph databases, such as:
• - Neo4j
• - Amazon Neptune
• - Memgraph
• - Search technologies, including:
• - Hybrid Search
• - BM25
• - Dense Retrieval
• - Sparse Retrieval
• - Semantic Search
• - Metadata Filtering
• - Re-ranking
• - Knowledge Graph integration
• LLM Evaluation
• - Experience designing systematic evaluation frameworks using tools such as:
• - Ragas
• - TruLens
• - DeepEval
• - OpenAI Evals
• - LangSmith Evaluation
• - Understanding of:
• - Hallucination detection
• - Faithfulness
• - Answer relevancy
• - Context precision
• - Context recall
• - Groundedness
• - Toxicity
• - Regression testing
• Guardrails & Observability
• - Guardrails:
• - Guardrails AI
• - NeMo Guardrails
• - OpenAI Moderation
• - Prompt Injection Detection
• - PII masking
• - Content filtering
• - Observability:
• - LangSmith
• - Langfuse
• - Arize Phoenix
• - Weights & Biases
• - MLflow
• - Experience with:
• - Prompt tracing
• - Token analytics
• - Cost monitoring
• - Latency monitoring
• - User feedback loops
• - Production debugging
• Prompt Engineering
• - Expertise in:
• - Chain-of-Thought (CoT)
• - ReAct
• - Tree of Thoughts
• - Self-Consistency
• - Few-shot prompting
• - Structured prompting
• - Function Calling
• - JSON mode
• - Prompt optimization
• - Prompt caching
• - Context window optimization
• - Token usage optimization
• - Cost optimization
• Model Deployment & Inference
• - Hands-on experience deploying open-source LLMs.
• - Preferred models:
• - Llama
• - Mistral
• - Qwen
• - Gemma
• - DeepSeek
• - Inference engines:
• - vLLM
• - TensorRT-LLM
• - Ollama
• - TGI (Text Generation Inference)
• - SGLang
• - Experience with:
• - GPU optimization
• - Batch inference
• - Model serving
• - Autoscaling
• - Multi-GPU deployment
• - Quantization (GGUF, GPTQ, AWQ, FP8, INT8, INT4)
• MLOps / AI Platform
• - Experience with:
• - MLflow
• - Kubeflow
• - Docker
• - Kubernetes
• - GitHub Actions / GitLab CI
• - Model versioning
• - Experiment tracking
• - Feature stores
• - Continuous evaluation
• - Continuous deployment
• Cloud Platforms
• - Experience with one or more:
• - Google Cloud Platform (Vertex AI)
• - Microsoft Azure AI
• - AWS Bedrock
• - OpenAI Azure
• Database Technologies
• - Experience with:
• - PostgreSQL
• - Oracle
• - MongoDB
• - Redis
• - Elasticsearch / OpenSearch
• Soft Skills
• - Strong analytical and problem-solving skills.
• - Excellent communication and stakeholder management.
• - Ability to lead technical discussions and architecture reviews.
• - Experience mentoring engineering teams.
• - Ability to work in Agile delivery environments.
• Banking & Industry Experience (Nice to Have)
• - Banking or Financial Services domain experience.
• - Experience with enterprise AI governance and Responsible AI frameworks.
• - Knowledge of SAMA, NCA, or other financial regulatory environments.
• - Experience building AI copilots and enterprise AI assistants.
• - Knowledge of OCR, document intelligence, and intelligent automation.
• - Experience integrating AI solutions with BPM/workflow platforms such as Appian, Camunda, or Pega.
• Education & Certifications
• - Bachelor's degree in Computer Science, Artificial Intelligence, Information Technology, Engineering, or a related field.
• - The following professional certifications are an advantage:
• - Google Professional Machine Learning Engineer
• - Microsoft Azure AI Engineer Associate
• - AWS Certified Machine Learning – Specialty
• - Databricks Machine Learning Professional
• - NVIDIA AI Certifications
• - OpenAI or Anthropic ecosystem certifications (where applicable)
• Key Competencies
• Agentic AI | Multi-Agent Systems | Large Language Models (LLMs) | RAG | Vector & Graph Databases | LLM Evaluation | Guardrails & Observability | Prompt Engineering | LLM Deployment & Inference | MLOps | Python | APIs & Microservices | Cloud AI Platforms | Responsible AI

Responsibilities

Skills

• Python programming
• RESTful API development
• FastAPI
• Agentic AI
• Multi-agent systems
• Retrieval-Augmented Generation (RAG)
• Large Language Models (LLMs)
• Embeddings and vector databases (Pinecone, Qdrant, Milvus, Weaviate, ChromaDB)
• Knowledge graphs (Neo4j, Amazon Neptune, Memgraph)
• LLM evaluation frameworks (Ragas, TruLens, DeepEval, OpenAI Evals, LangSmith Evaluation)
• Guardrails & Observability tools (NeMo Guardrails, LangSmith, Langfuse, Arize Phoenix, Weights & Biases, MLflow)
• Prompt engineering (CoT, ReAct, Tree of Thoughts, Self-Consistency)
• Model deployment and inference (Llama, Mistral, Qwen, Gemma; vLLM, TensorRT-LLM, Ollama, TGI)
• GPU optimization and multi-GPU deployment
• MLOps / AI Platform