Agentic AI Engineer
Job description
About the role
This role designs and ships autonomous agentic AI for tax and compliance workflows. The team builds customer-facing solutions that operate reliably in regulated financial environments. Success depends on combining AI innovation with strict accuracy and compliance.
Software engineers turn product ideas into working code. Engineers work in small teams, review each other's work, and ship in small batches. Most teams follow agile practices such as sprints and daily standups. Engineers also write tests, fix bugs, and improve performance. The field values clear communication as much as technical skill. Engineers spend part of every week on planning, code review, and debugging, not just writing new code. The ability to explain a technical decision in plain words separates strong engineers from the rest.
Key facts
What you'll do
Outputs maintain accuracy and dependability for Taxbit's enterprise customers across complex financial data.
The team resolves compliance and tax automation challenges through carefully scoped AI capabilities.
Reliability, accuracy, and user trust in financial workflows measure the success of these outcomes.
Well-designed, well-tested, and maintainable code supports long-term maintainability.
The team iterates based on behavior and feedback to keep systems aligned with compliance and performance standards.
Technical decisions and validation practices for AI-driven tax workflows reflect compliance requirements.
Requirements
3+ years of professional software engineering experience, with at least 1 year building and shipping AI or LLM-powered applications in production. This background ensures familiarity with deployment challenges and production AI operations.
Hands-on experience with AWS AI and ML services, including Bedrock, Strands, AgentCore, SageMaker, or equivalent agentic frameworks. These services support scalable and secure agentic system implementations for financial workloads.
Strong grasp of agentic patterns such as tool use, RAG, memory management, multi-agent orchestration, and human-in-the-loop design. Mastery of these patterns improves reliability in compliance-sensitive scenarios.
Designing and evaluating LLM pipelines, including prompt engineering, output validation, and hallucination mitigation, maintains factual accuracy. These practices support trust in automated tax reporting.
Familiarity with AI observability, evaluation frameworks, and responsible AI practices is essential for monitoring and improving production systems. Such practices guide debugging and optimization of production AI.
Agile collaboration communicates complex AI concepts clearly to technical and non-technical audiences.
A Bachelor's degree in Computer Science, Machine Learning, a relevant technical field, or equivalent practical experience demonstrates applied problem-solving. Practical experience often shows real-world system proficiency.
Practical notes
The role is based in Seattle, Washington, with a hybrid model of 3 days in-office and 2 days WFH or flexible work. The team operates with high ownership and autonomy, expecting strong initiative from engineers. Good to know
Agentic AI systems coordinate multiple tools and steps to complete tasks with minimal human intervention. These systems rely on clear goals, memory, and tool use.
Trustworthy AI in compliance domains demands rigorous validation, monitoring, and guardrails to prevent errors in financial outputs. Reliability and correctness are central to user confidence.
Modern AI engineering for finance uses cloud-native infrastructure, observability pipelines, and iterative evaluation to balance innovation with safety. Teams work closely with compliance and product stakeholders to align technical solutions to regulatory requirements.
Well-designed prompts and agent workflows incorporate feedback loops, versioning, and testing to maintain consistent behavior across diverse inputs and edge cases. Continuous refinement keeps systems aligned with evolving business rules.
Experience with AWS serverless and AI services supports scalable, secure, and cost-effective deployment of agentic workflows in production environments. These services reduce operational overhead while enabling rapid experimentation.