English for Edtech Developers

Edtech engineering blends education theory with product engineering, and its English vocabulary reflects that mix — terms like “scaffolding” and “formative assessment” come from pedagogy, while “LMS integration” and “adaptive engine” are pure software concepts. Being fluent in both halves helps you communicate clearly with instructional designers, teachers, and product stakeholders who may not share your engineering background.

Key Vocabulary

LMS (Learning Management System) A platform that hosts, delivers, and tracks educational content and learner progress — such as Canvas, Moodle, or a custom-built system. Example: “We’re building a plugin so our course content can be embedded directly inside the school’s existing LMS.”

LTI (Learning Tools Interoperability) A standard protocol that allows external educational tools to integrate securely with an LMS, handling authentication and grade passback. Example: “We implemented LTI 1.3 so teachers can launch our tool directly from their LMS gradebook without a separate login.”

Adaptive learning A system that adjusts the difficulty, sequence, or type of content presented to a learner based on their ongoing performance, rather than following a fixed path for everyone. Example: “The adaptive engine bumped this student to more advanced exercises after three correct answers in a row on the current skill.”

Formative vs. summative assessment Formative assessment checks understanding during the learning process to guide instruction (like a quick quiz); summative assessment evaluates learning at the end of a unit (like a final exam). Example: “This in-lesson check is formative — it shouldn’t affect the student’s grade, just flag which concepts need review.”

Scaffolding Instructional support provided to a learner that is gradually reduced as they become more capable — in software, often implemented as hints, worked examples, or partial solutions that fade over time. Example: “The exercise starts with a worked example, then removes scaffolding step by step until the student solves the final problem unaided.”

Item bank A repository of individual assessment questions (“items”), often tagged by skill, difficulty, and standard alignment, used to assemble quizzes or adaptive assessments dynamically. Example: “We need to expand the item bank for this skill — there are only four questions, so the adaptive engine keeps repeating them.”

Rubric A structured scoring guide that defines specific criteria and performance levels for evaluating open-ended student work, such as an essay or project. Example: “The auto-grading feature applies the teacher’s rubric criteria, but flags any submission below a confidence threshold for manual review.”

Mastery-based progression A learning model where students advance to new material only after demonstrating proficiency in the current material, rather than progressing purely by time spent or seat time. Example: “Under mastery-based progression, a student can’t unlock the next module until they score above 80% on the current one.”

Common Phrases

In code reviews:

• “This grade passback call isn’t retrying on failure, so if the LMS is briefly unavailable, the teacher’s gradebook silently falls out of sync.”
• “The adaptive engine is recalculating difficulty on every single answer — can we batch this to avoid whiplash in question difficulty?”
• “This accessibility check is missing alt text validation for the diagram-heavy questions, which is a real barrier for screen reader users.”

In standups:

• “Yesterday I finished LTI grade passback; today I’m working on the item bank tagging so questions can be filtered by standard alignment.”
• “I’m blocked on the rubric editor — the product team wants weighted criteria, and our current data model assumes equal weighting.”
• “I fixed a bug where mastery progression let a student skip ahead after a single lucky guess instead of requiring sustained performance.”

In meetings with instructional designers or educators:

• “When you say ‘this should feel more scaffolded,’ can you give an example of the kind of hint you’d want at step two?”
• “Should this be a formative check that doesn’t affect the grade, or should it count toward the final score?”
• “The adaptive engine currently treats all wrong answers the same — should certain misconceptions trigger different remediation content?”

Phrases to Avoid

Saying “the test” for every kind of assessment. Educators distinguish carefully between “a formative check,” “a quiz,” “a summative assessment,” and “a diagnostic.” Using “test” indiscriminately can create real confusion about grading stakes and how the result should be used.

Saying “the AI grades it” without describing the actual mechanism. Say instead: “the system applies the teacher’s rubric automatically” or “an automated scorer suggests a grade that the teacher can review and override” — this matters both for setting expectations and for accuracy in describing the product to stakeholders and users.

Saying “make it more personalized” without specifying the mechanism. “Personalized” is used loosely across edtech. Clarify whether you mean adaptive difficulty, content recommendation, pacing adjustment, or interface customization — each requires very different engineering work.

Quick Reference

Key Takeaways

• Distinguish formative from summative assessment precisely — the distinction affects grading, UX, and stakes for the learner.
• Use “personalized” and “adaptive” carefully; specify the actual mechanism (difficulty adjustment, content recommendation, pacing) rather than the marketing term alone.
• Scaffolding and mastery-based progression are core pedagogy concepts that should shape how you describe hint systems and progression logic.
• When discussing AI-assisted grading, be precise about the human’s role — describe it as suggesting or supporting, not replacing, teacher judgment.
• LMS/LTI integration vocabulary is essential for any edtech engineer working with schools — know grade passback and launch flows well enough to discuss failure modes clearly.