Full-Stack AI Engineer (.NET + React)

We are looking for a strong Full-Stack .NET + React Engineer with knowledge in AI LLM integrations to help expand and improve an existing AI CFO product. The MVP is already built and in use. This role is focused on building new features, improving AI Agent performance, and developing new Agents.

You will work directly on an active product and help us:

• Design and implement new features on top of an existing system
• Build advanced file export and reporting functionality (PDF, Excel, structured reports)
• Improve accuracy, consistency, and performance of LLM outputs
• Maintain and optimize generating charts and graphs using LLMs
• Create evaluation pipelines to measure and optimize model performance
• Implement validation layers and deterministic post-processing

Core Requirements

Full-Stack Engineering

• React (or similar frontend framework) experience
• .NET Backend development experience
• Working with AWS (or similar cloud environments)
• Knowledge with relative and document databases 

AI / LLM Engineering

• Proven experience integrating LLMs into real applications
• Experience with MCP and understanding how it works
• Structured output generation (JSON, reports, analytics)
• Prompt engineering and workflow optimization
• Building evaluation frameworks and test suites for AI systems
• Experience with frontier LLM models
• Handling retries, validation, guardrails, and fallback logic
• Optimizing cost, latency, and accuracy trade-offs
• Use AI-assisted coding tools (Claude Code) to ship faster

Strong Plus

• Financial analytics or accounting-related software experience
• Familiarity with QuickBooks or similar financial systems

Engagement Model

• Part-time contract (~120 hours per month)
• Fully remote
• Flexible schedule
• Long-term collaboration (not a short gig)
• Complete freedom in choosing technologies and architectural solutions

This is perfect for an experienced engineer who prefers flexible contract work and enjoys building practical AI systems.

Shortlisted candidates will receive a test task. Detailed project information will be shared after signing an NDA.