linkedin
Founding AI Product Engineer
Argentic Technologies Inc.
Abu Dhabi, UAE
Full Time
Senior
4 weeks ago
LLM powered SystemsAgentic WorkflowsStructured ExtractionWorkflow OrchestrationEval HarnessesRetrieval Strategies
Free
Job Fit Check
Base Career helps you apply smarter for this job.
?%
Ready to ScanKey skills for this role
LLM powered SystemsAgentic WorkflowsStructured Extraction
About the Role
Argentic Technologies seeks a founding AI Product Engineer to build agent-first systems for B2B payments. The role involves designing and building systems that turn messy payment workflows into controlled, auditable outputs using LLMs, deterministic logic, and human-in-the-loop review.
Key Skills for This Role
LLM powered SystemsAgentic WorkflowsStructured ExtractionWorkflow OrchestrationEval HarnessesRetrieval Strategies
Responsibilities
- Build agent first workflows for payment intake, remittance capture, matching, exception review, and ERP ready output
- Combine LLMs, deterministic business logic, workflow state, structured extraction, human review, and audit trails
- Create eval harnesses to measure whether AI workflows are actually improving finance operations
- Experiment with models, prompts, tool use, retrieval strategies, structured output, and orchestration patterns
- Ship product directly with founders, customers, and design/product context close at hand
- Work across backend, integrations, data workflows, internal tools, and product infrastructure
- Help decide what should be automated, what should be reviewed by a human, and what should remain deterministic
Requirements
- You have built LLM powered or agentic systems beyond simple chat interfaces
- You can explain tradeoffs across model quality, latency, cost, reliability, tool use, and structured output
- You understand why evals matter and have built or used harnesses to compare prompts, models, retrieval strategies, or workflows
- You are comfortable turning messy business processes into working technical systems
- You care about reliability, observability, failure modes, review paths, and auditability
- You can move between prototype speed and production discipline
- You are curious about new AI capabilities, but practical enough to know when simpler code is better than another agent
Full Job Posting
Role Overview
- Argentic is looking for an early engineering hire to help build agent first systems for the messy receive side of B2B payments.
- This is not a generic full stack role, and it is not a prompt wrapper role. We are looking for someone deeply hands on with modern AI systems: models, tools, agents, evals, retrieval, workflow orchestration, structured extraction, and human in the loop review.
- Argentic is building a supplier side B2B payments platform.
What you will work on
- Building agent first workflows for payment intake, remittance capture, matching, exception review, and ERP ready output
- Combining LLMs, deterministic business logic, workflow state, structured extraction, human review, and audit trails
- Creating eval harnesses to measure whether AI workflows are actually improving finance operations
- Experimenting with models, prompts, tool use, retrieval strategies, structured output, and orchestration patterns
- Shipping product directly with founders, customers, and design/product context close at hand
- Working across backend, integrations, data workflows, internal tools, and product infrastructure
- Helping decide what should be automated, what should be reviewed by a human, and what should remain deterministic
What we are looking for
- You have built LLM powered or agentic systems beyond simple chat interfaces
- You can explain tradeoffs across model quality, latency, cost, reliability, tool use, and structured output
- You understand why evals matter and have built or used harnesses to compare prompts, models, retrieval strategies, or workflows
- You are comfortable turning messy business processes into working technical systems
- You care about reliability, observability, failure modes, review paths, and auditability
- You can move between prototype speed and production discipline
- You are curious about new AI capabilities, but practical enough to know when simpler code is better than another agent
Relevant experience
- Relevant experience could come from fintech, payments, banking, accounting, treasury, AP, AR, payroll, tax, commerce, vertical SaaS, workflow automation, document extraction, reconciliation, data operations, finance operations, or integration heavy products.
- You do not need to be a payments expert on day one, but you should be excited by operational complexity where correctness matters.
This may not be the right role if
- Your AI experience is limited to calling a model API inside a standard app flow
- You see agents as magic rather than systems that need state, tools, constraints, tests, and review paths
- You need clean requirements before making progress
- You only want to work on frontend polish or only on infrastructure
- You prefer optimizing inside a mature engineering organization over helping build the engineering culture from the ground up
Apply for this job in 1 click
Skip the repetitive application forms
Install the Base Career Chrome Extension and autofill job applications across major job boards with your profile.
Trusted by over 500,000 job seekers on Base Career