Artificial Intelligence Engineer

Overview

Join Luxoft as an expert Artificial Intelligence Engineer tasked with designing and implementing large language model (LLM) applications and autonomous agent workflows on Azure cloud platforms. You will lead the development of RAG pipelines, multi-agent orchestration, integration of MCP tools, and maintain high standards for generative AI engineering across projects, ensuring quality meets defined evaluation thresholds.

Key Responsibilities

• Develop LLM applications and agent workflows using LangGraph, LangChain, or Semantic Kernel on Azure OpenAI, with additional exposure to AWS Bedrock or GCP Vertex advantageous.
• Architect Retrieval-Augmented Generation (RAG) pipelines involving document ingestion, effective chunking, embeddings creation, and vector search technologies like Azure AI Search, FAISS, or Pinecone. Implement hybrid retrieval and ranking solutions.
• Integrate enterprise systems via Model Context Protocol (MCP) servers, facilitate function and tool calls, and orchestrate multi-agent environments with robust state management.
• Build, version, and rigorously test prompts, conducting both offline and live evaluations focusing on accuracy, grounding, hallucination rates, and apply fine-tuning when necessary.
• Implement responsible AI safeguards including content filtering, PII redaction, policy compliance checks, and LLM observability features such as tracing, token usage tracking, and latency monitoring.
• Optimize model inference and token consumption costs through smart techniques like model routing, caching, and batching while monitoring adherence to cost budgets.
• Document prompt designs, evaluation outcomes, and model selection rationale. Provide mentoring to team members adopting generative AI patterns.

Required Qualifications and Skills

• Bachelor's degree in Computer Science, Engineering, Information Systems, or related disciplines, or equivalent relevant experience.
• Over 10 years of experience in software or data engineering, including at least 3 years in machine learning or natural language processing, and a minimum of 2 years delivering production-grade LLM solutions leveraging Azure OpenAI or similar platforms.
• Advanced proficiency in Python programming, experience with LangGraph, LangChain, or Semantic Kernel frameworks, and expertise in asynchronous, event-driven software architectures.
• Deep knowledge of Azure OpenAI services and related Azure AI components such as Azure AI Search, AI Foundry/ML, Azure Functions, Container Apps/AKS, Key Vault, and Application Insights.
• Expertise designing RAG workflows encompassing chunking strategy, embedding generation, vector database utilization (Azure AI Search, FAISS, Pinecone), hybrid search, and re-ranking mechanisms.
• Development of MCP servers and clients, implementation of function calling, and proficient multi-agent orchestration with effective state handling.
• Experience conducting LLM evaluation using tools like RAGAS, promptfoo, Azure AI evaluation or equivalent, alongside fine-tuning methods such as LoRA and PEFT, with informed model selection.
• Implementation of responsible AI guardrails using Azure AI Content Safety, NeMo Guardrails, or similar frameworks; LLM observability utilizing LangSmith, OpenTelemetry; and financial operations management for inference cost control.

Preferred Additional Experience

• Familiarity with AWS Bedrock, GCP Vertex AI, semantic caching, model distillation, and deployment of small language models.
• Industry certifications such as Azure AI Engineer Associate (AI-102) or equivalent credentials.
• Background in financial services, sovereign wealth funds, investment holdings, or similarly regulated enterprise sectors.
• Experience collaborating with distributed teams, notably working onsite in UAE alongside nearshore and offshore teams in India and Poland.

Language Proficiency

Advanced English skills at C1 level.