Prompt Engineer
Job description
About the role
Netomi is actively seeking a Prompt Engineer to join the team and take ownership of the end to end prompt lifecycle for client solutions. In this capacity, you will design, implement, and iteratively refine prompts that directly influence the accuracy, reliability, and performance of AI driven systems. A core part of this role involves translating complex business rules and nuanced customer requirements into structured prompt instructions that guide large language models toward desired outcomes. You will be responsible for ensuring that every prompt is not only technically sound but also contextually aligned with client objectives and operational constraints. This position requires a deep understanding of how AI behavior can be shaped through careful linguistic and structural design. You will work in close collaboration with cross functional partners to ensure that the solutions you build are practical, scalable, and maintainable in production environments. The work you do will directly impact the quality of automated customer interactions and the efficiency of issue resolution. This role is positioned at the intersection of software engineering, data science, and domain expertise, requiring a methodical and detail oriented approach to problem solving.
Key facts
What you'll do
Develop comprehensive prompt architectures that address specific client use cases and business requirements.
Define standardized tool descriptions and function signatures within agentic frameworks to streamline AI understanding and execution.
Analyze and enhance existing prompt structures to improve clarity, efficiency, and overall system performance.
Automate the execution of prompt testing routines using robust scripts to ensure consistency and repeatability in evaluations.
Conduct in depth evaluations of multiple large language models to determine the best fit for performance, cost, and reliability criteria.
Construct and maintain evaluation frameworks and benchmark datasets to establish industry standard best practices for prompt engineering.
Collaborate daily with Customer Success and Data Science teams to align prompt strategies with evolving product roadmaps and support needs.
Create and update detailed documentation that captures prompt development workflows, optimization techniques, and decision rationales.
Monitor the latest research and practical advancements in natural language processing to inform prompt design decisions.
Experiment with novel prompting strategies, few shot techniques, and chain of thought reasoning to solve complex problems.
Adapt foundation models to meet specific client constraints, domain terminology, and compliance requirements.
Perform root cause analysis when AI responses deviate from expectations, adjusting prompts to correct behavior and reduce errors.
Ensure that all prompt related artifacts are version controlled, reviewed, and archived for future reference and auditability.
Support the deployment of prompts into staging and production environments, working closely with engineering operations to ensure smooth integration.
Requirements
Hold a Bachelor's or Master's degree in Computer Science, Data Science, Linguistics, or a closely related technical field.
Possess a solid and demonstrable grasp of natural language processing concepts, including tokenization, embeddings, and model architectures.
Have a firm understanding of machine learning fundamentals such as training workflows, loss functions, and evaluation metrics.
Bring prior, verifiable experience in AI prompt engineering, including the full cycle of writing, optimizing, and evaluating prompts for real world applications.
Be skilled in programming, with a strong focus on Python and its ecosystem for data manipulation and automation.
Show familiarity with AI frameworks and libraries such as TensorFlow, PyTorch, and Hugging Face, and be able to apply them in practical scenarios.
Demonstrate excellent analytical and problem solving capabilities when diagnosing complex model behavior and prompt failures.
Exhibit strong written and verbal communication skills, with the ability to explain technical concepts to both technical and non technical stakeholders.
Be capable of working effectively within a collaborative team environment while independently managing several projects and priorities under tight deadlines.
Maintain a high attention to detail and a disciplined approach to testing, documentation, and quality assurance.
Demonstrate intellectual curiosity and a commitment to continuous learning in a rapidly evolving technological landscape.
Show resilience and adaptability when working with ambiguous requirements and iterative feedback cycles.
Possess strong time management skills to meet deliverables and support timely client engagements.
Exhibit professionalism and integrity in all interactions with colleagues, partners, and clients.
Nice to have
Bring at least 5 years of hands on experience in data science, with a track record of managing complex data pipelines and experiments.
Have direct experience working with large language models such as GPT 3, GPT 4, and Llama 3, including fine tuning, prompt tuning, and inference optimization.
Show proficiency with data visualization tools and methods to communicate model performance and prompt effectiveness.
Have a background in linguistics or computational linguistics, with an understanding of syntactic and semantic analysis.
Experience with agile development methodologies, including sprint planning, standups, and retrospectives, is highly valued.
Practical notes
This is a full time position based in Toronto, Canada.
Applicants must be authorized to work in Canada without sponsorship.
Visa sponsorship is not available for this role.
Relocation support is not provided.
The selected candidate is expected to commit to the full duration of the engagement.
Applications will be reviewed on a rolling basis until the position is filled.
Candidates are encouraged to apply promptly to ensure full consideration.
All hired employees must comply with Netomi policies and standard employment requirements.