USD per year
Applied LLM Engineer
Location
San Francisco, CA, US
Job Type
Full-time
Role
Engineering, Backend
Experience Level
3+ years
Visa Requirement
US citizen/visa only
Skills
- JavaScript
- Python
- React
- Amazon Web Services (AWS)
- FastAPI
- Django
- Terraform
- Pulumi
- GitHub Actions
- LLM-based applications development
- Multi-step LLM workflows design
- Task-specific agents creation
- Experience with frontier models (OpenAI, Anthropic, Google)
- AI code editors usage
- Prompt engineering strategies development
- Evaluation frameworks and RAG pipelines implementation
- Technical R&D in model functionality boundaries exploration
- Secure coding practices in regulated industries (life sciences, healthcare, fintech)
- Frontend development experience (React) - plus but not required
Company Info: Artos
- Founded: 2023
- Batch: W24
- Team Size: 7
- Status: Active
- Mission: Build tools that help biopharma companies create and manage their R&D documentation in a fraction of the time.
- Product supports companies from innovative biotech startups to large pharmaceutical companies to deliver life-saving treatments faster.
- Uses AI to create key documents like FDA submissions for life sciences companies.
About the Role
At Artos, you will be a core member of the engineering team playing a critical role in developing, scaling, and expanding the Artos platform to serve regulatory needs for pharma and life science companies globally.
Qualifications
- Bachelor’s or Master’s degree in Computer Science, Engineering or related field (or equivalent practical experience)
- 3+ years software development experience focused on building and deploying AI/ML applications
- Strong backend engineering experience including:
- Building APIs using Python frameworks such as FastAPI and Django
- Deploying and scaling containerized applications in cloud environments (AWS, GCP, Azure)
- Implementing CI/CD pipelines (e.g., GitHub Actions)
- Working with Infrastructure-as-Code tools such as Terraform or Pulumi
- Hands-on experience building LLM-based applications including:
- Designing multi-step LLM workflows and task-specific agents
- Experience with frontier models (OpenAI, Anthropic, Google etc.)
- Experience with AI tools as a user (AI code editors)
- Developing advanced prompt engineering strategies evaluation frameworks and RAG pipelines
- Conducting technical R&D to explore model functionality boundaries
- Familiarity with secure coding practices ideally in regulated industries (life sciences healthcare fintech)
- Experience working in or adjacent to regulated domains (life sciences clinical R&D) is a plus
- Frontend development experience (React) is a plus but not required
Requirements / Responsibilities
- Design and maintain scalable production-grade backend systems for AI applications
- Create orchestrate and evaluate LLM-based agents and chained workflows with minimal oversight
- Debug and improve LLM-driven systems identifying issues across model output API behavior system logic layers
- Conduct rapid experimentation and research on LLM capabilities translating findings into production functionality
- Stay current with emerging practices models tooling in generative AI ecosystem and apply pragmatically
- Communicate clearly with technical and nontechnical collaborators (product managers medical writers customer teams)
- Operate effectively in fast-paced ambiguous environment managing shifting priorities novel problem spaces
Other Information / Work Environment
- Comfortable working in fast-paced intense startup environment
- Willingness to work in-person at office in Mission Bay 4–5 days per week
Founder Contact / Team Lead for Role
Josh Kim Founder
Substrate AI provides AI Native Revenue Cycle Management (RCM) software that autonomously completes mission-critical tasks in healthcare revenue cycle such as claim status, appeals, policy reviews, payment posting, refunds, and more. Their agents connect to healthcare systems and run accounts receivable autonomously from claim status and appeals to payment posting and refunds. The software integrates with PM, EHR, and clearinghouse systems without migration and is live within 48 hours for most practices. It offers features like full audit trail, human-in-the-loop escalation for uncertain cases, predictive insights on payer trends and denial patterns, eligibility checks, payer policy review, document review automation, and tailored automation pipelines built by AI engineers embedded with client teams. The platform emphasizes security with end-to-end encryption, zero-trust architecture, periodic security audits, granular access controls, and compliance with HIPAA and SOC2.
View Company Profile