Educational AI Engineer: Personalize Learning with AI Prototypes

Weld North Education (now Imagine Learning)

Boston (MA)

Hybrid

USD 129,000 - 200,000

Full time

42 hours ago

Be an early applicant

A complete application in a minute

Tailored resume and cover letter, ready to send.

Get past ATS filters

Benefits offered by this job

Health, dental, and vision plans

401k with company match

Paid holidays

Paid time off

Tuition reimbursement

Work-from-home opportunities

Professional development programs

Travel opportunities

Job summary

Imagine Learning seeks an AI Engineer to join a small, entrepreneurial R&D team. You will prototype AI applications, work with model APIs and tools, and translate ideas into reliable software for educational outcomes.

Requirements include 5+ years in software/ML, strong Python/TypeScript, and familiarity with sandboxing and security in agentic systems. The role offers hybrid work in the Boston area with travel as needed.

Qualifications

Bachelor’s degree in computer science, engineering, or related field; 5+ years software or ML engineering experience.
Hands-on experience building LLM applications with tool use, retrieval, structured outputs, and evaluation.
Strong Python or TypeScript skills; experience with APIs, asynchronous code, data models, testing, cloud services, and debugging.
Understanding of sandboxing, isolation, access controls, secrets management, and security risks in agentic systems.
Preferred: experience with multimodal or voice AI, agent orchestration, model adaptation, or evaluation infrastructure.
Preferred: experience with agentic toolkits/frameworks like Deep Agents, LangChain, LangGraph, OpenClaw, Hermes.

Responsibilities

Prototype AI applications and agentic workflows using model APIs and tool calling.
Compare models and architectures; apply fine‑tuning or adaptation as needed.
Build evaluation datasets, automated checks, and human-review workflows.
Diagnose failures with traces and error analysis; improve grounding and tool reliability.
Design sandboxed execution with least-privilege access and security controls.
Test prompt injection, data leakage, and unintended actions; ensure auditable execution.
Measure latency and throughput; collaborate with engineers and researchers on testable software.
Document findings, share learnings, and help decide next steps.
Perform other duties as required.