Promega Instructional Design Internship · May-August 2026

Adaptive Learning Assistant

I proposed an AI-supported learning experience, scoped the build for roughly ten weeks, and developed a working prototype with fixed routing rules and a course-grounded mentor.

Problem framingScope decisionsPrototype testing
4

Modules in the pilot course

~10

Weeks for the prototype build

3

Instructional-design colleagues in the internal pilot

Working internship prototype · Internally tested with instructional designers · No target-learner impact data

Product perspective · View learning-design perspective →

Pain Point

An opportunity to focus training on existing knowledge gaps.

Promega's Scientific and Sales Training Services (SSTS) team supports global employees and distributor teams with product and sales training. I proposed exploring whether assessing prior knowledge could let learners focus on gaps and avoid repeating familiar content. I discussed the idea with my supervisor and developed a structured plan. Greater efficiency and personalization were design hypotheses; the internal pilot did not measure those outcomes.

My Role

Scope decisions and constraints

A working prototype within roughly ten weeks

Use fixed tests instead of computer-adaptive testing

I researched computer-adaptive testing, then chose fixed pre- and post-test question sets to fit the available time. They covered the same course topics with different questions, using the team's existing 80% standard.

Keep routing decisions outside the AI

Concerns about hallucinations led me to keep scoring, module assignment, and completion in deterministic rules. The AI mentor provided explanations from SME-reviewed course summaries and refused assessment-answer requests.

Find a workable way to deliver existing content

A Storyline hosting attempt through Supabase did not work. I explored alternatives with Claude and integrated an existing Rise course through Review 360. This made the course accessible, although opening it separately created friction between study and mentor support.

Record the delivery boundary

The built prototype stored profiles, scores, progress and timestamps, and displayed learner-facing summaries. Sending post-test results to a manager or SME remained planned and was not implemented.

Learning Technology

Two systems shared one learner profile.

The prototype paired a rules-based engine with a course-grounded AI mentor. That division let the product use generative AI where it helped learning, while keeping assessment, routing, and completion decisions explainable.

Shared Learner Profile
  • Learner intake: name, role, and year at Promega
  • Assessment record: pre- and post-test scores
  • Course state: the current module each learner is on
  • Module timestamps: when learners open and complete content
Rule-Based EngineGrades assessments, routes modules, and determines completion.

I designed the assessment logic and per-module mastery rules so routing decisions stayed consistent and auditable.

Course-Grounded AI MentorExplains concepts, coaches learners, and protects assessment answers.

I used reviewed course content summaries to guide the AI, configured refusals for test-answer requests, and flagged out-of-scope questions for human support. The flag did not send a message to an expert.

Prototype Walkthrough

How the learning decisions appear to learners

These screens show the learner journey from intake to course completion and how the design appeared in the web application.

Learner setup and diagnostic screens
01

Create a profile and diagnose prior knowledge

Learners enter their name, role, and years at Promega to log in and access the course. Their progress is saved so they can return later after logging in again. The pre-course check then identifies what they already know.

Learner intake creates the profile used throughout the experience.
Pre-course assessment with capillary electrophoresis questions and answer options.
The diagnostic maps 20 questions to four course modules.
02

Turn assessment evidence into a learning plan

Immediate answer feedback supports transparency, while module-level scores determine which content is required and which becomes optional review.

Pre-test answer review showing correct answers and explanations.
Question-level feedback explains both correct and incorrect responses.
Pre-test summary showing module scores, mastery status, and assigned learning plan.
This summary at the end of the pre-test shows the accuracy score for each module. The 80% mastery rule then converts those results into required and optional modules.
03

Focus learning on identified gaps

The course panel keeps required modules, progress, records, and mentor support together. The existing Rise course opens separately through Review 360; learners return to the app for mentor support and assessment.

Course panel showing required and optional modules, locked post-test, records, and AI mentor.
The main workspace makes the assigned path and post-test requirements visible.
Introduction to Capillary Electrophoresis module displayed in Review 360.
Existing Rise course content supplied the instruction; I built the surrounding learning path and support.
04

Make AI behavior visible and accountable

Every mentor response is labeled so learners can see whether it is grounded in the course, redirected away from test answers, or flagged for human support.

AI mentor response labeled Grounded in the course.
Course-grounded explanation.
AI mentor response redirecting a request for post-test answers.
Assessment-answer protection.
AI mentor response flagged for a human because the question is outside the course scope.
Human-support flag for out-of-scope questions; no automatic expert notification.
Post-test access after required modules
05

Gate the post-test by required work

The system unlocks the post-test only after all assigned modules are complete. The mentor recognizes the learner's progress and remains available for review.

Course panel with required modules completed, post-test unlocked, and mentor coaching visible.
Completion state, post-test access, and mentor context stay synchronized.
06

Respond to both post-test outcomes

The system branches at the 80% mastery threshold. Learners who pass receive confirmation and completion records. Learners below the threshold receive review guidance and a summary. The screen described an intended human follow-up, but sending results or notifications to a manager or SME was not implemented.

Pass pathway · 80% or higher

Post-test passed screen showing a 90 percent score, answer feedback, and mentor response.
The pass state combines assessment evidence with continued mentor support.
Post-test summary with module scores, mentor summary, and record downloads.
Passing learners receive module scores, a mentor summary, and downloadable records.

Support pathway · Below 80%

Post-test result below the 80 percent threshold with a supportive review message.
The learner sees a clear, supportive message that the mastery threshold was not met.
Post-test summary showing a 55 percent score, modules to review, mentor guidance, and downloadable records.
Module-level results guide review. Manager/SME report delivery remained planned.

Testing & Insights

Internal testing checked the workflow and surfaced design issues.

I wrote a tester guide and asked three instructional-design colleagues to try the prototype on their own computers and challenge the chatbot with difficult questions. They supplied written feedback about the workflow, answers, and usability. The core flow worked in this exploratory pilot. There was no time to implement the findings below before the internship ended.

The prototype worked end to end, including module access, learner persistence, and mentor responses.

Assessment quality needed refinement because some questions felt guessable and answer lengths sometimes hinted at the correct option.

Learners needed clearer guidance around what to do next, when they were done, and how the mentor should be used.

Internal testers encountered the intended refusal behavior, but strict refusals sometimes created friction. This pilot did not establish reliability across all possible questions.

Opening course modules outside the main app made it harder to use the mentor while studying.

Project Scope

As a prototype, the project showed feasibility before learner-impact measurement.

What this project demonstrated

The prototype showed that adaptive routing, guarded AI support, learner persistence, and completion reporting could work together in a single learning experience.

What would need more evidence

This project is a prototype and has not been applied to real learners. There is no target-group data yet, so learning-impact claims would need a structured learner pilot.

Assessment design

The routing rule depends on the quality of the questions.

I consolidated a content inventory with an instructional designer using her existing Claude workflow. An SME helped identify essential knowledge. I then used Claude to produce pre- and post-test drafts covering the same topics with different questions, with five questions per module, and coordinated review by two SMEs.

Translate evidence into a path

The prototype applies the team's existing 80% module threshold: four correct answers out of five make that module optional review. Lower scores assign the module as required content. The course content itself was an existing Articulate Rise course accessed through Review 360.

Recognize the limits of the rule

One answer changes a module score by 20 percentage points. Pilot feedback about guessable questions therefore matters to routing accuracy. The threshold is a prototype rule, not evidence that the assessment predicts workplace performance.

Explanatory feedback for each answer

At my supervisor's suggestion, I implemented prepared feedback for each answer option. A correct choice received an explanation of why it was correct. An incorrect choice received an explanation of the mistake and the correct answer. SMEs also reviewed these explanations.

Keep review findings separate from completed revisions

Reviewers found generally credible content, but some answers were guessable because of option length or implausible distractors. Review arrived late in the project, and those suggested assessment revisions were not implemented before the internship ended.

Learning Design

Assess first, route deliberately, support on demand.

Assessment results determine the path. The AI mentor supports explanations without deciding which modules a learner can skip.

  1. Assess prior knowledgePre-test results are recorded by module.
  2. Assign a learning pathFixed rules mark modules required or optional.
  3. Practice with supportExisting course content and the AI mentor support study.
  4. Review the post-testLearners see scores, explanations, and review guidance.

Automatic delivery of results to a manager or SME remained planned. The prototype did not implement that handoff.

Reflection

Keeping learning decisions explainable.

I kept assessment scoring, module routing, and completion in fixed rules, while using AI for course-grounded explanations and support. Building the experience made the quality of the assessment questions especially visible: a working rule still depends on the evidence used to apply it.