Back to portfolio
Content Generation · HR/People

Expert Instructional Designer

Instructional Design Advisor GPT

Unstructured training ideas → structured with proven learning frameworks → clearer, more effective instructional content.

01The Problem

Creating effective training content requires more than subject expertise; it demands structure, pedagogy, and audience alignment. Without this, learning materials are often disorganized, disengaging, or misaligned with learner needs. This leads to confusion, low retention, and ineffective training outcomes.

02What the AI Does

I generate, structure, evaluate, and refine instructional content using established frameworks like ADDIE, SAM, and Bloom’s Taxonomy. I guide users through audience analysis, learning objective design, content sequencing, assessment creation, and engagement strategies. I am built on a GPT-5.3 language model, configured via a detailed system prompt that: * Constrains me to instructional design expertise * Directs me to provide actionable, pedagogy-grounded recommendations * Encourages alignment with learning science principles * Emphasizes clarity over jargon I do not access external tools, databases, or proprietary knowledge bases. My outputs are shaped entirely by prompt instructions and general training.

03Design Decisions

01 · Choice

Narrow specialization in instructional design

Why

To produce higher-quality, domain-specific guidance instead of generic responses [Creator: add rationale]

Constraint

Limits scope to learning design problems; avoids drifting into unrelated general advice

02 · Choice

Explicit grounding in established frameworks (ADDIE, SAM, Bloom’s Taxonomy)

Why

To anchor outputs in widely accepted methodologies rather than ad hoc suggestions

Constraint

Ensures recommendations follow structured, pedagogically sound approaches

03 · Choice

Instruction to tailor content to audience characteristics (age, prior knowledge, modality)

Why

Effective learning depends heavily on audience alignment [Creator: add rationale]

Constraint

Prevents one-size-fits-all outputs; pushes toward contextualized recommendations

04 · Choice

Emphasis on actionable, clear guidance over theory-heavy explanations

Why

Users likely need usable outputs rather than academic discussion [Creator: add rationale]

Constraint

Reduces abstraction; prioritizes practical application

05 · Choice

Built-in guidance on engagement, assessment, and feedback loops

Why

Instructional design is not just content creation but includes evaluation and iteration

Constraint

Ensures outputs address full learning lifecycle, not just content drafting

06 · Choice

Inclusion of accessibility and inclusivity considerations

Why

Modern instructional design standards require inclusive practices [Creator: add rationale]

Constraint

Encourages outputs that consider diverse learners and accessibility needs

07 · Choice

No tool or data integration (pure prompt-based system)

Why

Simplicity and portability across contexts [Creator: add rationale]

Constraint

Cannot validate against real learner data or LMS analytics

08 · Choice

Tone calibration to be supportive but not overly praising or vague

Why

Maintain credibility and usefulness for professional users

Constraint

Avoids fluff; focuses on substantive feedback and critique

05Key Insight

AI becomes significantly more useful when constrained by domain frameworks that shape not just what it generates, but how it thinks.