AI Personalized Learning Specialist
An AI Personalized Learning Specialist designs, implements, and optimizes AI-driven systems that create adaptive, individualized l…
Skill Guide
The systematic discipline of designing, structuring, and optimizing natural language inputs (prompts) and interactive instruction sequences to reliably elicit specific, high-quality outputs from Large Language Models (LLMs) and shape them into functional, pedagogically sound AI tutoring agents.
Scenario
You need to create a bot for a SaaS company that answers FAQs, referencing specific help articles without hallucinating.
Scenario
Create an AI tutor that doesn't give direct answers to coding problems but guides a student to discover the solution through targeted questions and hints.
Scenario
Build a system where an AI mediator guides a human trainee through a negotiation with one or more AI-powered counterparts, each with distinct hidden objectives and personalities.
Core platforms for prompt iteration and API integration. Use API playgrounds for rapid prototyping, and LangChain/LlamaIndex for complex orchestration involving memory, retrieval, and chains. PromptFlow is for enterprise-grade deployment and monitoring.
CRISPE (Capacity, Role, Insight, Statement, Personality, Experiment) provides a structured template for complex prompts. The refinement cycle (Design -> Test -> Analyze Failure -> Redesign) is mandatory. Treat prompts like code in Git to track changes and performance regressions.
Answer Strategy
The answer must demonstrate a shift from single prompts to a system. Strategy: Describe a stateful architecture. Sample answer: 'I'd implement a multi-turn architecture with a core pedagogical prompt and a separate, hidden 'assessment analyzer' prompt. The tutor prompt's instructions would reference a dynamic student profile. After each interaction, the analyzer would evaluate the student's last response for correctness and depth, updating the profile with an estimated mastery level (e.g., 0.7). The tutor prompt would then use this variable to choose its next action-simplifying an analogy if mastery is low, or introducing a harder problem if it's high.'
Answer Strategy
Tests debugging and safety mindset. Strategy: Show systematic process and prioritization. Sample answer: 'First, I'd immediately roll back to the last known stable prompt version. Next, I'd run the failing queries through a log analysis to identify the new failure pattern. The issue likely stems from a recent change to the prompt's 'instruction' or 'persona' section that removed explicit constraints on length and risk. I'd fix it by adding direct, imperative constraints like "Be concise" and "Do not speculate on legal implications. If unsure, state you lack information and escalate to a human." Then, I'd add these specific failing cases to my regression test suite.'
1 career found
Try a different search term.