Numeral
Software Engineer (AI)
About this role
Numeral is seeking a Foundational AI Engineer to develop and maintain AI-driven automation systems within a rapidly growing team. This role involves designing production systems, implementing AI integrations, and collaborating closely with other teams to enhance user experience.
What you'll do
- Design and implement production AI systems
- Maintain AI integrations using modern frameworks
- Build scalable backend services and APIs
- Design automated workflows and orchestration systems
- Make architectural decisions focused on performance and reliability
- Define engineering best practices and technical culture
What they're looking for
- 7+ years in product development
- 2+ years with LLMs
- Experience with PostgreSQL and Redis
- Proficiency in front-end technologies like TypeScript
- Strong product sense and problem-solving skills
- Excellent communication and collaboration
- Ability to thrive in dynamic startup environments
- Experience with automation tools
Benefits
- Competitive salary and equity
- Comprehensive medical, dental, and vision coverage
- Wellness perks such as Headspace and Peloton One App
- 401(k) for long-term financial security
- Lunch and snacks in-office
- Opportunities for team offsites and company events
Opens the application — the Jobs AI extension fills it for you. Set up autofill
Opens the official application on the employer’s site. No login required.
Numeral
Numeral builds an AI-driven tax platform that automates and streamlines tax processes through intelligent systems. The company is hiring Solutions Engineers, Software Engineers, Full-Stack Engineers, and Foundational AI Engineers to develop scalable infrastructure, implement AI automation, and support customer success.
View all jobs at NumeralLikely interview questions
- Walk us through a production LLM system you've built end-to-end. What were the key architectural decisions around model orchestration, latency, and cost?
- Describe your experience with tool-calling agents or agentic workflows. How did you handle reliability and error recovery in production?