About the role
As an Applied LLM Engineer at TheWebVale, you will turn language model capabilities into production software that solves business challenges. You will construct patient intake triage tools, legal clause extractors, and automated student feedback generators for client organizations.
You will collaborate with backend developers and project managers in our standard two-to-three week delivery cycles. You will ensure applications balance accuracy, response speed, and operational costs while handing full IP to our clients.
What you will do
- Build practical software applications that apply language models to real business tasks.
- Implement high-accuracy extraction pipelines for unstructured invoices, receipts, and forms.
- Design few-shot prompt libraries that steer models toward consistent business formatting.
- Implement caching and batching strategies to minimize API token expenditure for clients.
- Connect model outputs directly to relational database tables and transaction records.
- Write comprehensive automated unit tests covering typical and edge-case text inputs.
What you will need
- Hold 2–4 years of software development experience with applied LLM implementation.
- Demonstrate strong proficiency in Python or TypeScript with asynchronous API coding.
- Show practical experience integrating OpenAI, Anthropic, or Google generative APIs.
- Understand token budgeting, prompt caching, structured schema outputs, and rate limits.
- Possess disciplined coding practices with Git version control and automated testing.
- Communicate project progress and technical boundaries clearly to non-technical founders.
Nice to have
- Experience with Instructor or Outlines libraries for structured output generation.
- Familiarity with document parsing tools like PyMuPDF, pdfplumber, or Marker.
- Knowledge of vector databases and semantic search integration.