Purpose of the role
Design, build and operationalise production- grade Generative AI and Large Language Model (LLM) solutions that unlock business value for Peterson Solutions and its clients—e.g., intelligent document processing, knowledge assistants and automated content generation. The role owns the full solution lifecycle (research → prototype → production), ensures reliability and cost- effectiveness in Azure, and collaborates closely with product, data and DevOps teams to deliver measurable impact.
Responsibilities
Track and improve model KPIs (accuracy, latency, hallucination rate) through continuous experimentation and A/B testing.
Produce high- quality technical documentation and conduct knowledge- sharing sessions for engineers and non- technical stakeholders.
Partner with Product Owners to translate business requirements into technical roadmaps and deliverables.
Develop prompt schemas, function calling and agent workflows with Semantic Kernel or LangChain.
End- to- end ownership of Generative AI features—from feasibility study, model selection and prompt/agent design to production roll- out and monitoring.
Establish robust MLOps practices: Docker packaging, CI/CD, automated testing, observability dashboards and cost optimisation.
Integrate Azure OpenAI, Azure AI Search or other LLM endpoints into the existing micro- service architecture (REST/GraphQL).
Architect and implement Retrieval- Augmented Generation (RAG) pipelines: data ingestion, embedding creation and vector search.
Essential
Bachelor’s degree (or higher) in Computer Science, Software Engineering, Data Science or a related field.
≥ 3 years’ experience in machine learning/NLP, including ≥ 1 year hands- on with LLMs or Generative AI.
Deep understanding of LLM architectures, prompt engineering, fine- tuning and parameter- efficient techniques.
Demonstrated success in deploying an AI solution to production.
Excellent analytical, problem- solving and communication skills; proficient English (spoken & written).
Practical expertise with Semantic Kernel or LangChain and at least one vector database (Azure AI Search, Pinecone, Weaviate, etc.).
Strong coding proficiency in C or Python / Node.js.
Solid grasp of software- engineering best practices: Git, design patterns, containerisation, Agile delivery.
Familiarity with cloud AI stacks—Azure strongly preferred; AWS/GCP equivalents a plus.
Nice- to- Have
Experience with Intelligent Document Processing (Azure Form Recognizer, LayoutLM, OpenCV).
Past deployment of chatbots or knowledge assistants to Microsoft Teams, Slack or web channels.
Knowledge of responsible- AI guardrails (content moderation, data privacy, bias mitigation).
Contributions to open- source GenAI projects or published research.
