Prompt Engineer
JD:
We are seeking a highly skilled and detail-oriented Prompt Engineer to design, test, and refine prompts for large language models to generate high-quality, executable code across multiple programming languages. As a Prompt Engineer, you will work closely with our AI research and development teams to optimize the interaction between LLMs and code generation, ensuring that the outputs meet specific functionality, syntax, and quality requirements.
Key Responsibilities:
- Prompt Design & Engineering: Develop, test, and iterate on LLM prompts to generate accurate, syntactically correct code in various programming languages (e.g., Python, JavaScript, SQL).
- Contextual Programming: Ensure that generated code aligns with the user’s intent and incorporates contextual information such as function inputs/outputs, documentation, and architecture patterns.
- Code Optimization: Identify and resolve issues with LLM-generated code, including syntax errors, inefficiencies, and performance bottlenecks.
- AI Model Fine-tuning: Collaborate with the research team to improve model
- Cross-Functional Collaboration: Work closely with the larger team to develop tools and frameworks for code generation workflows.
- Continuous Improvement: Research and implement best practices for prompt engineering and explore new techniques to improve LLM performance in code generation tasks.
Key Qualifications:
- Proven experience with prompt engineering and LLMs like GPT-3, GPT-4, Codex, or other advanced models.
- Strong background in software development with expertise in languages such as Python (Required), JavaScript, SQL, C++, Java, or others.
- Deep understanding of ML, NLP, and LLMOps principles.
- Proficiency in code debugging, optimization, and test automation for generated code.
- Experience with software version control tools (e.g., Git)
- Strong problem-solving skills and the ability to think critically about how to improve LLM performance for specific use cases.
- Excellent collaboration and communication skills, with the ability to explain complex technical topics to diverse teams.
Preferred Qualifications:
- Experience with template-driven code generation frameworks such as Jinja or Velocity
- Familiarity with using LLMs in DevOps environments.
- Experience working with AI/ML development environments and libraries such as Hugging Face, OpenAI API, or LangChain.