Role overview
Position Name: AI Engineer
Reports to: DevOps Manager
Location/Type: UA/Remote
Atlas Technica's mission is to shoulder IT management, user support, and cybersecurity for our clients, who are hedge funds and other investment firms. Founded in 2016, we have grown year over year through our uncompromising focus on service.
We value ownership, execution, growth, intelligence, and camaraderie. We are looking for people who share our Core Values, thrive, and contribute to this environment while putting the customer first. At Atlas Technica, we offer a competitive salary, comprehensive benefits, and great perks to our global Team. We strive to maintain a professional yet friendly environment while promoting professional and career development for our Team Members. Join Atlas Technica now!
As an AI engineer, you will design and deploy intelligent chatbots using Foundry Workflows, Microsoft Agent Framework, OpenAI; build and operate multi-agent orchestration pipelines with Azure AI Foundry and Model Context Protocol (MCP); integrate AI with enterprise systems via RESTful APIs, ensuring compliance with security and privacy standards. Stay current with LLM advancements, collaborate across teams, and mentor junior engineers.
What you'll work on
- AI Agent & Chatbot Development
- Design, build, and deploy conversational agents using Azure AI Foundry, Microsoft Agent Framework, and OpenAI.
- Develop prompt strategies for context-aware, multi-turn dialogue.
- Design agentic loops with task decomposition, state passing, and workflow enforcement.
- Implement structured error responses and graceful degradation in agent tool calls.
- Data & Model Engineering
- Build and maintain data ingestion and indexing pipelines for Azure AI Search and Cosmos DB.
- Evaluate and optimize AI system performance for accuracy, scalability, latency, and cost.
- System Integration & Architecture
- Integrate AI with enterprise systems (ConnectWise PSA, Confluence) via RESTful APIs.
- Implement MCP servers for secure, typed tool integration between agents and enterprise APIs.
- Deploy containerized services to Azure using multi-stage builds and CI/CD pipelines.
- Security, Compliance & Governance
- Ensure AI systems comply with data privacy regulations (GDPR) and security standards.
- Implement access controls, encryption, and audit logging for AI workflows.
- Research & Innovation
- Stay current with LLM technologies, frameworks, and methodologies.
- Evaluate emerging tools to improve solution quality and delivery speed.
- Collaboration & Leadership
- Work with cross-functional teams to recommend LLM-driven solutions.
- Contribute to architectural decisions aligned with strategic goals.
- Mentor junior engineers and support knowledge sharing.
- Operational Excellence
- Own incident resolution and bug fixes.
- Create and maintain technical documentation for AI systems and integrations.
What we're looking for
- English level – B2 or higher
- 5+ years of experience in AI engineering and.NET software development (.NET 10, ASP.NET Core/WebAPI, C#) with frontend experience in React 19 / Next.js 15 / TypeScript, with a focus on Azure-based solutions.
- Experience building intelligent chatbots and autonomous agents (Agentic AI) that perform real actions, plan, reason, and act with minimal human input; familiar with multi-agent orchestration and agent memory.
- Strong expertise in embeddings, search, and prompt engineering.
- Strong understanding of large language models (LLMs), their limitations, and modern capabilities (e.g., GPT, Claude, Gemini).
- Hands-on experience with Azure AI services (Azure AI Foundry, Azure OpenAI, Azure AI Search) and hands-on experience with MCP or similar protocols for secure, scalable integration between AI agents and enterprise systems.
- Familiarity with multi-agent orchestration patterns: coordinator-subagent workflows, maker-checker validation, signal-based confidence scoring, and context preservation across multi-turn interactions.
- Experience designing prompts with explicit evaluation criteria, few-shot techniques, structured output via JSON schemas, and validation/retry loops for reliable LLM responses.
- Awareness of AI-specific security concerns: PII handling, data boundary enforcement, prompt injection prevention, and compliance considerations (e.g., GDPR).
- Ability to write clean, maintainable, and testable code following SOLID principles and Clean Architecture patterns.