Project description
Luxoft is initiating the development of a solution designed to generate investment insights based on sales and research materials. The solution will leverage advanced Agentic AI capabilities to significantly reduce the time required to prepare for client meetings and improve quality of the insights.
Responsibilities
Lead the architecture and implementation of production-grade LLM orchestration, multi-agent systems, and autonomous coding workflows
Oversee the development of enterprise Retrieval-Augmented Generation (RAG) pipelines, semantic chunking strategies, and vector database integrations
Implement strict evaluation and observability frameworks to monitor production latency, API costs, system drift, and model accuracy
Drive the design, deployment, and ongoing maintenance of secure, scalable, and resilient cloud systems on AWS leveraging ECS Fargate, Lambda, SQS, Aurora, and Neptune
Own the end-to-end infrastructure lifecycle by embedding robust Infrastructure-as-Code (IaC) and deployment pipelines directly within the development workflow
Lead, mentor, and inspire a specialized team of Python and TypeScript engineers
Drive delivery utilizing agile methodologies, balancing rapid AI prototyping with enterprise-grade stability, security, and compliance
Act as the primary technical liaison between business stakeholders, product managers, and the engineering team to translate strategic goals into technical realities
Skills
Must have
10+ years of professional software development experience.
Advanced proficiency in Python (asyncio, FastAPI) and TypeScript (Next.js/React, serverless execution layers)
Hands-on experience building complex, stateful agentic workflows using LangGraph or LangChain
Proven track record architecting, provisioning, and managing your own production infrastructure on AWS, specifically utilizing ECS Fargate, Lambda, and SQS
Experience defining cloud architecture programmatically using advanced Infrastructure-as-Code (IaC) tools like AWS CDK or Terraform (Python/TypeScript preferred)
Experience managing relational databases (Aurora) alongside graph or vector backends (Neptune)
Power-user fluency with advanced command-line AI interfaces (Claude Code CLI, GitHub Copilot CLI) with a deep understanding of prompt engineering and context window management
Seasoned technical leader with professional software engineering experience and a proven track record leading teams to work collectively to deliver solutions on time
Experience operating in an agile setting, deploying and maintaining AI/LLM applications in a live, enterprise-scale production environment
Accountable for results, with excellent communication skills to mentor engineers and defuse technical friction
Nice to have
Highly desirable: experience with AgentCore, AWS Neptune, Amazon API gateway
Familiarity with ML fundamentals relevant to content generation (embeddings, tokenization, evaluation, fine tuning/LoRA, prompt+retrieval evaluation).
Other
Languages
English: B2 Upper Intermediate
Seniority
Lead
