What You'll Do You’ll join the HS Code Classification team, which helps businesses that ship goods across borders get the right tariff code for every product, and cuts down the manual work our classification experts do today. Here you’ll own models from first experiment to production, work on a hard real-world problem with thousands of codes and messy product data, and see the results used directly by customers. Build intelligent systems that classify, extract, validate, and reason over complex business data and multiple languages Improve automation quality through applied ML, GenAI, retrieval, and evaluation-driven development Integrate model improvements into production workflows and user-facing experiences Support fast, reliable ML services that can operate in real-time and high-volume environments Improve data quality detection so downstream AI systems operate on inputs that are complete and actionable You’ll work on text classification, data quality checks, model evaluation, small language models, RAG, and faster inference. You’ll also build AI agents that can reason through multi-step tasks, build MCP to connect them to tools and data they need. The role needs hands-on ML skills, strong engineering, and the ability to collaborate across product, engineering, and domain teams. This role is remote. #LI-Remote What Your Responsibilities Will Be Build and improve ML systems for product classification, document understanding, data quality checks, and retrieval Fine-tune, test, and deploy models, including transformers and small language models Build AI agents that reason through multi-step tasks and use MCP to connect to tools and data Write backend services, APIs, and pipelines that bring models into production Set up evaluations to track accuracy, speed, reliability, and where models fail Add GenAI features such as chat, retrieval, and clear explanations of results to the product Keep systems production-ready with testing, monitoring, and documentation Collaborate with product managers, engineers, and domain experts to turn ambiguous needs into working solutions Take part in design reviews, code reviews, and experiments to keep improving our systems Main tools: Python, transformer models, LLMs, MCP, Docker, Kubernetes, MLflow, AWS, GitLab, Postgres, and Prometheus, Hex Dashboards , CI/CD deployments What You'll Need to be Successful Bachelor’s or master’s degree in Computer Science, Engineering, or a related technical field. 5+ years of industry experience in software engineering, including building and deploying ML or LLM-based systems in production. Strong Python skills, plus experience building backend services, APIs, and distributed systems. Familiarity with other languages is a plus. Hands-on experience building applications with LLMs, such as prompt engineering, tool and function calling, retrieval, structured outputs, and working with models via APIs or Bedrock. Practical experience with agentic development, including multi-step agents, orchestration patterns, and agent frameworks. Delivered production-ready ML improvements for one or more AI-powered workflows. Improved model quality, latency, reliability, or user experience for a production ML system Built or enhanced evaluation pipelines that make model performance easier to measure and improve Integrated ML capabilities into backend services, APIs, or user-facing workflows Contributed to data quality or input validation capabilities that improve downstream model performance Demonstrated strong ownership of well-scoped technical workstreams with guidance from senior engineers Improved engineering quality through reliable implementation, testing, documentation, and production support. Avalara is an AI-first Company AI is embedded in our workflows, decision-making, and products. Success here requires embracing AI as an essential capability. You’ll bring experience using AI and AI-related technologies, ready to thrive here. You’ll apply AI every day to business challenges - improving efficiency, contributing solutions, and driving results for your team, our company, and our customers. You’ll grow with AI by staying curious about new trends and best practices, and by sharing what you learn so others can benefit too. How We'll Take Care of You Total Rewards In addition to a great compensation package, paid time off, and paid parental leave, many Avalara employees are eligible for bonuses. Health & Wellness Benefits vary by location but generally include private medical, life, and disability insurance. Inclusive culture and diversity Avalara strongly supports diversity, equity, and inclusion, and is committed to integrating them into our business practices and our organizational culture. We also have a total of 8 employee-run resource groups, each with senior leadership and exec sponsorship. What You Need To Know About Avalara We’re defining the relationship between tax and tech. We’ve already built an industry-leading cloud compliance platform, processing over 54 billion customer API calls and over 6.6 million tax returns a year. Our growth is real - we're a billion dollar business - and we’re not slowing down until we’ve achieved our mission - to be part of every transaction in the world. We’re bright, innovative, and disruptive, like the orange we love to wear. It captures our quirky spirit and optimistic mindset. It shows off the culture we’ve designed, that empowers our people to win. We’ve been different from day one. Join us, and your career will be too. We’re An Equal Opportunity Employer Supporting diversity and inclusion is a cornerstone of our company — we don’t want people to fit into our culture, but to enrich it. All qualified candidates will receive consideration for employment without regard to race, color, creed, religion, age, gender, national orientation, disability, sexual orientation, US Veteran status, or any other factor protected by law. If you require any reasonable adjustments during the recruitment process, please let us know. Build and improve ML systems for product classification, document understanding, data quality checks, and retrieval Fine-tune, test, and deploy models, including transformers and small language models Build AI agents that reason through multi-step tasks and use MCP to connect to tools and data Write backend services, APIs, and pipelines that bring models into production Set up evaluations to track accuracy, speed, reliability, and where models fail Add GenAI features such as chat, retrieval, and clear explanations of results to the product Keep systems production-ready with testing, monitoring, and documentation Collaborate with product managers, engineers, and domain experts to turn ambiguous needs into working solutions Take part in design reviews, code reviews, and experiments to keep improving our systems Main tools: Python, transformer models, LLMs, MCP, Docker, Kubernetes, MLflow, AWS, GitLab, Postgres, and Prometheus, Hex Dashboards , CI/CD deployments Bachelor’s or master’s degree in Computer Science, Engineering, or a related technical field. 5+ years of industry experience in software engineering, including building and deploying ML or LLM-based systems in production. Strong Python skills, plus experience building backend services, APIs, and distributed systems. Familiarity with other languages is a plus. Hands-on experience building applications with LLMs, such as prompt engineering, tool and function calling, retrieval, structured outputs, and working with models via APIs or Bedrock. Practical experience with agentic development, including multi-step agents, orchestration patterns, and agent frameworks. Delivered production-ready ML improvements for one or more AI-powered workflows. Improved model quality, latency, reliability, or user experience for a production ML system Built or enhanced evaluation pipelines that make model performance easier to measure and improve Integrated ML capabilities into backend services, APIs, or user-facing workflows Contributed to data quality or input validation capabilities that improve downstream model performance Demonstrated strong ownership of well-scoped technical workstreams with guidance from senior engineers Improved engineering quality through reliable implementation, testing, documentation, and production support.
