Agentic AI Engineer
Taxbit offers enterprise-grade tax and accounting tools for digital assets, automating information and financial reporting through a modern, compliant platform.
Investors
About Taxbit
Taxbit provides an enterprise-grade tax and accounting platform designed for digital assets. The platform automates information and financial reporting to help businesses stay compliant with global regulations. Key products include Information Reporting, which automates the ingestion of data and generation of required tax forms, and Financial Reporting, which offers a subledger solution that integrates with existing systems to categorize transactions, generate financial reports, and update general ledgers. Taxbit serves a wide range of industries, including digital asset brokers, traditional brokers, centralized exchanges, DEXs, payment processors, NFT marketplaces, and more, and is trusted by Fortune 500 companies and government entities.
Skills
About the Role
You'll be at the forefront of applied AI, designing and shipping agentic systems that operate autonomously, reason across complex data, and deliver world-class experiences to enterprise customers. You'll design, build, and deploy customer-facing agentic AI products using AWS-native technologies, architect multi-agent systems with reliable tool use, memory, and orchestration patterns, and partner with product managers and customer success to translate user problems into elegant, production-grade AI solutions. You'll own the full lifecycle of AI features, from prompt engineering and agent design through evaluation, deployment, and monitoring, while writing well-designed, well-tested, and maintainable code and participating in code reviews. You'll instrument and monitor agentic systems in production and collaborate cross-functionally with engineers, data teams, and compliance stakeholders to ensure AI outputs meet accuracy and regulatory standards.
Requirements
- 3+ years of professional software engineering experience, with at least 1 year building and shipping AI/LLM-powered applications in production
- Hands-on experience with AWS AI/ML services - Bedrock, Strands, AgentCore, SageMaker, or equivalent agentic frameworks
- Strong understanding of agentic patterns: tool use, RAG, memory management, multi-agent orchestration, and human-in-the-loop design
- Experience designing and evaluating LLM pipelines including prompt engineering, output validation, and hallucination mitigation
- Familiarity with AI observability, evaluation frameworks, and responsible AI practices
- Ability to work in an agile environment and communicate complex AI concepts clearly to both technical and non-technical audiences
- Bachelor's degree in Computer Science, Machine Learning, relevant technical field, or equivalent practical experience
- Strong coding skills in Python and/or TypeScript
- Experience with AWS cloud infrastructure, IAM, Lambda, and related services; DataDog and GitHub CI/CD a plus
Responsibilities
- Design, build, and deploy customer-facing agentic AI products using AWS-native technologies including Strands Agents, AgentCore, Bedrock, and Lambda
- Architect multi-agent systems with reliable tool use, memory, and orchestration patterns that perform accurately in high-stakes financial contexts
- Partner with product managers and customer success to understand user problems and translate them into production-grade AI solutions
- Own the full lifecycle of AI features from prompt engineering and agent design through evaluation, deployment, and monitoring
- Write well-designed, well-tested, and maintainable code and participate in code reviews
- Instrument and monitor agentic systems in production using observability tooling and iterate rapidly based on real user behavior and feedback
- Collaborate cross-functionally with engineers, data teams, and compliance stakeholders to ensure AI outputs meet accuracy and regulatory standards
Benefits
- Equity (RSUs)
- Competitive benefits package
- A modern 401(k) plan that includes access to crypto, financial wellness benefits, low fees and more
- Hybrid working model: 3 days in-office, 2 days WFH/flexible
- Monday team lunches, snacks and drinks
- Discretionary Time Off
- Paid parental leave
- Fertility Benefit
- Autonomous work and flexibility in how work is performed