Senior AI Application Developer

At Signature Aviation we are modernizing operations and customer experience through advanced data platforms and artificial intelligence. We are hiring an Senior AI Application Developer to build the AI capabilities the business actually uses — agentic systems, LLM-powered applications, and intelligent automation on Azure — together with the retrieval, feature, and data pipelines those capabilities depend on. You own the capability end to end: the agent or assistant, the data it reasons over, the pipelines that feed it, and the evaluation that proves it works. You will not hand off at the data boundary, and you will not be handed a clean dataset.

Working model — forward deployed: you embed with the business unit and the engineering teams that own a use case, build inside their environment and release process, hand back a running system, and contribute what you learned to the shared accelerator library between deployments. 

Qualifications:

Minimum Education and/or Experience

• 5 years in software, data, or ML engineering, with recent hands-on delivery.

• You have built and deployed production-grade LLM applications or agent-based AI systems, and can explain how one worked, what broke, and how you knew it was good enough to ship.

• Azure is required: hands-on Microsoft Foundry including Azure OpenAI models — deployments, the Foundry Models catalog, and Azure AI Content Safety.

• Cloud-native application patterns on Azure: containers on Azure Container Apps or AKS, Azure Functions, CI/CD through Azure DevOps or GitHub Actions, Microsoft Entra ID managed identity, and Key Vault-backed configuration.

• LLM application patterns — prompt design, RAG, function and tool calling, structured outputs, streaming — plus at least one orchestration framework: the Microsoft Agent Framework, the Foundry Agent Service, Semantic Kernel, LangChain/LangGraph, or LlamaIndex.

• Demonstrated data engineering: you have built or substantially rebuilt ingestion, transformation, embedding, or feature pipelines and kept them correct in production.

• Vector or hybrid search in Azure AI Search, Cosmos DB, or pgvector; Pinecone, Weaviate, or Redis Vector considered as equivalent depth.

• You have evaluated model or prompt quality systematically. A lightweight harness you built yourself counts; anecdotal assessment does not.

• Strong Python and working TypeScript.

• Experience integrating AI systems with enterprise data and operational platforms.

• You have worked directly with the people who use what you build: embedded delivery, solutions engineering, or stakeholder-facing product work.

• Sound instincts on secrets, access control, and PII, and the judgment to stop and ask.

• Bachelor's degree in Computer Science, Engineering, or a related technical field, or equivalent practical experience.

Preferred Qualifications

•  Strong LangGraph and multi-agent orchestration experience.

•  MCP and enterprise tool integration.

•  Microsoft Foundry, Azure OpenAI, Azure AI Search, Azure Databricks or Microsoft Fabric.

•  OpenAI, Claude, Gemini, Llama or other foundation models.

•  Langfuse, LangSmith, Ragas, Promptfoo or Azure AI Evaluation SDK.

•  MLflow and model lifecycle experience.

•  Experience with coding agents such as Claude Code, OpenAI Codex or GitHub Copilot.

• Experience working in a Forward Deployed Engineer, solutions engineering, consulting, startup or internal innovation role.

• Aviation, logistics, mobility or other large-scale operational environments.

Additional Information

This is an applied engineering role, not a research role: building statistical and forecasting models sits with data science, and you build the systems that put their outputs to work.

Responsibilities:

Essential Duties and Responsibilities
• Build and deploy production LLM applications, agentic systems and multi-agent solutions using LangGraph, Microsoft Agent Framework, Semantic Kernel or similar frameworks.
• Design stateful agent workflows using tool calling, conditional routing, checkpointing, structured outputs, human-in-the-loop, memory and failure recovery; understand when a single agent, multi-agent system or deterministic workflow is the better approach.
• Integrate agents securely with enterprise systems through APIs, function calling and MCP, with appropriate authentication, authorization, approval and audit controls.
• Apply context engineering across prompts, retrieved information, conversation state, tool results, business rules and short- or long-term memory.
• Build RAG and retrieval capabilities using Azure AI Search, Databricks Vector Search, pgvector, Cosmos DB or equivalent technologies, including chunking, embeddings, hybrid search, metadata filtering and reranking.
• Build and maintain the ingestion, transformation, embedding and feature pipelines required by AI capabilities using Azure Databricks, Microsoft Fabric or equivalent platforms.
• Build systematic evaluation using golden datasets, automated scoring, LLM-as-judge and human review; evaluate accuracy, grounded-ness, hallucination, task completion, agent trajectories, safety, latency and cost.
• Instrument model calls, retrieval, agent steps and tool calls using Microsoft Foundry tracing, OpenTelemetry, Langfuse, LangSmith or equivalent tools, and support testing for prompt injection, PII leakage and excessive agent autonomy.
• Work as a Forward Deployed Engineer with business and engineering teams, taking use cases from an ambiguous problem through experiment, production deployment and operational handoff.
• Turn reusable components and lessons from each deployment into shared AI patterns, libraries and accelerators, and work with the AI Platform Engineer to follow enterprise standards for security, model access, evaluation, observability and cost.