AI Engineer

Company Description

Alyra Technology builds AI-enabled software and automation systems for small and mid-sized businesses, with a focus on document-heavy, regulated environments. We combine modern cloud engineering with applied AI to help teams work faster, reduce manual effort, and deliver consistent, verifiable outcomes.

Role: AI Platform Engineer (Multi-Agent Document Search Platform) - Contract (20–40 hrs/week, Remote)

We're hiring a Full Stack Engineer to help build a secure, production-grade platform that ingests complex documents and powers a multi-agent search system capable of reasoning, retrieving, and generating traceable, auditable insights.

This role fits someone strong in React, experienced with Azure cloud architecture, and capable of building multi-agent systems that maintain detailed audit trails of system reasoning and data lineage. Familiarity with insurance policy or legal document structures is highly desirable. Our environment also includes Foundry as part of the workflow stack.

What You'll Do

Frontend (React)

• Build a React web app with authenticated user access, document upload, viewing, and search features.
• Create interfaces that display agent outputs alongside evidence panels, source references, and reasoning audit trails so users can understand how conclusions were formed.
• Design user experiences for processing status, version tracking, and clear error handling for long-running workflows.

Backend + Cloud (Azure)

• Develop scalable APIs (C# preferred; Python/Node acceptable) supporting ingestion, retrieval, and agent orchestration.
• Build and maintain Azure resources: Functions, App Services, Storage, Azure SQL/Postgres, Identity, Key Vault, and networking fundamentals.
• Own CI/CD pipelines (Azure DevOps) and ensure strong observability (logs, traces, metrics), safe deployments, and rollback patterns.
• Implement role-based access control and maintain a full audit history of user interactions and system events.

Document Ingestion

• Engineer pipelines to parse, normalize, chunk, and organize large document sets (PDFs and related formats).
• Store structured metadata to support document lineage and traceability from ingestion to agent output.
• Optimize throughput, cost, and accuracy with idempotent, fault-tolerant processing (retries, partial failures, reprocessing).

Multi-Agent Search and Reasoning

• Design and implement multi-agent orchestration with routing, specialized agent roles, and evidence synthesis.
• Log every step of agent reasoning, tool use, and retrieved evidence to produce user-visible audit trails.
• Add evaluation hooks and regression checks to maintain reliability across iterations.
• Ensure system outputs are source-backed, with clear linkage to the underlying documents.

Foundry Integration

• Contribute to workflow and deployment integrations involving Foundry as part of the platform stack.
• Ensure consistent permissions, versioning, and operational alignment between Foundry workflows and Azure services.

Minimum Qualifications

• 2+ years of experience building production software (backend services + APIs).
• Strong React experience building real product UI (auth, uploads, complex state management, error handling).
• Strong Azure experience (Functions, App Service, Storage, SQL/Postgres, Identity, Key Vault, networking fundamentals).
• Proficiency in C# for backend systems.
• Hands-on experience building multi-agent or agent-based systems beyond a toy demo.
• Experience implementing robust logging, monitoring, and auditability.
• Ability to own work end-to-end with minimal supervision.

Nice to Have

• Experience with insurance policy language, contract analysis, or structured legal/regulatory documents.
• Experience implementing agent reasoning transparency and audit trail visualization in a web app.
• Foundry experience in production environments.
• Experience with event-driven architectures (Service Bus, Event Grid, Kafka).
• Infrastructure as code (Terraform or Bicep).\
• Familiarity with modern AI developer tools and workflows, including Claude Code (or similar agentic coding tools), and strong comfort working alongside AI-assisted development in a production codebase.
• Experience working for consulting firms

Values and Working Style

• Drive and urgency: you move fast without creating chaos.
• Hardworking and dependable: you do what you say you'll do, and you finish.
• Ownership mindset: you take problems end-to-end and don't wait to be told.
• High standards: you care about correctness, security, and maintainability.
• Clear communication: you surface risks early and ask the right questions when requirements are ambiguous.
• Practical builder: you avoid over-engineering and ship what matters.

Please upload your resumes in English.