Sr .Net / AI Agents Developer

Before you apply, please get familiar with Luxoft

• Luxoft locations: https://career.luxoft.com/locations/
• Logeek Magazine: https://career.luxoft.com/logeek-magazine/
• Luxoft Alumni Club: https://career.luxoft.com/alumni/

Project Description:

• We’re looking for a self-directed engineer to help build the next generation of agentic systems at a leading global hedge fund. You’ll design and build systems that combine LLMs, software engineering, data, search, and domain-specific tools, and help shape how these systems reason, retrieve, and execute in production. This is an opportunity to solve hard engineering problems, work directly with investment professionals, and influence the evolution of a mission-critical platform.

Responsibilities:

• • Designing and building agent workflows, orchestration, and execution frameworks
• • Building retrieval and knowledge systems across structured and unstructured data
• • Evaluating and refining agent performance and behavior
• • Building distributed, production-grade services and APIs (Python/TypeScript, cloud infrastructure)
• • Evaluating new models, frameworks, and tools for production use
• • Investigating and resolving complex production issues across the stack
• This is a hands-on engineering role. You will prototype ideas, test them, break them, understand why they broke, and turn the useful ones into production systems.

Mandatory Skills Description:

• • 7+ years of software engineering experience
• • Strong Python and .NET skills
• • Hands-on skills in TypeScript and JavaScript
• • Solid CS fundamentals, including DSA
• • A track record of building distributed, service-oriented systems
• • Comfort with concurrent programming, APIs, messaging, and handling failures gracefully
• • Experience with Docker, containers, cloud infrastructure, and modern deployment practices
• • Knowledge of how AI agents are designed, including architecture, orchestration, and memory
• • Familiarity with retrieval and grounding techniques such as RAG, embeddings, and hybrid search
• • An understanding of how to evaluate, monitor, and improve agent behavior over time
• • A grasp of planning, tool and model selection, and sandboxed execution
• • An understanding of statistics and ML basics: how models, embeddings, inference, and evaluation work, and where probabilistic systems fall short. No ML research background needed.