Software Engineer 1, Full-Stack

Overview

The Software Engineer, Full-Stack develops, implements and documents software and programs that provide the technical solutions to meet specifications and business requirements defined by company objectives to promote effective, efficient, and compliant operations. This position builds end-to-end (full stack) features across LoopTech's AI products, including the TOPOS tutoring platform, from the user interface through APIs, data, and cloud deployment, using AI-first development workflows.

Responsibilities

Software Development & Delivery

• Work as part of a project team to define and document solutions for a variety of software development projects
• Contribute to all stages of the software development process, ensuring compliance with software development and architectural standards and achievement of documented requirements
• Use AI-first development workflows every day, with fluency in coding agents and harnesses (e.g., Claude Code, Codex) and their plugins, skills, extensions and MCP servers
• Review, verify and manage AI-generated code, agent loops and software factory pipelines, acting as the human in the loop: run the tests and check real behavior before merge
• Work within the existing and evolving agentic software development lifecycle (SDLC), in which agents are used throughout development
• Work test-first (TDD) within a CI/CD workflow, applying SOLID and other sound design principles
• Work in an Agile team, balancing research and development with production delivery and software quality

Frontend & Application Development

• Build and maintain frontend features for LoopTech web applications, such as course authoring, guided conversations, onboarding, and interactive visualizations
• Build with accessibility and mobile/tablet responsiveness as first-class requirements, not an afterthought
• Work with UX designers to turn Figma designs and interactive prototypes into production-ready, accessible, responsive interfaces, integrated with backend services, APIs and databases
• Validate end-to-end functionality across supported browsers, devices and screen sizes; troubleshoot integration issues and fix bugs
• Refine prototypes based on UX feedback, test results and technical requirements, keeping the application maintainable, performant and ready for deployment

Backend, Data & AI Services

• Design and extend REST API endpoints and backend services (in Python, Node.js and Rust) that power the student runtime, instructor tooling, and analytics/insights (per-subtopic and per-week mastery tracking)
• Design, migrate, and maintain databases (PostgreSQL with pgvector, SQLite) and blob/object file storage
• Work on the document ingestion pipeline, including large-file and multi-format uploads (PDF, PPTX, XLSX, CSV, HTML), chapter/outline splitting, deduplication, embedding generation, indexing, and vector search for retrieval-augmented generation (RAG), including graph RAG
• Extend these pipelines to build semantic graphs and to transcribe and summarize video and audio
• Integrate LLMs through APIs, from hosted providers to self-hosted open models, and real-time voice (LiveKit) to support live, voice-driven Socratic sessions

Security, Identity & Compliance

• Implement and harden authentication/authorization flows, including OIDC/SSO federation with university and customer login systems, Microsoft Entra ID, role-based access (student / instructor / account admin), and session management across web and mobile
• Support FERPA-aligned data handling, including personally identifiable information (PII), privacy-safe audit trails, data classification, and account/role separation, in partnership with the compliance workstream
• Keep secrets and student data out of AI tool context, and follow team rules for agent permissions
• Follow product safety rules for sensitive student content, such as crisis disclosures

Infrastructure & Release Engineering

• Contribute to infrastructure and release engineering, including Docker or Podman images, Terraform, Kubernetes/Helm deployments, CI/CD pipelines (GitHub Actions), and coordinated releases across Azure and customer cloud environments

Documentation, Maintenance & Support

• Bootstrap, develop, and maintain current-state documentation and deliverables for solutions
• Maintain existing and new software systems to ensure they continue to meet user needs
• Provide assistance to operational teams to ensure continuity of service
• Work with the Software Engineering community of practice to identify and implement continuous improvement opportunities

Qualifications

Required

• Bachelor's degree in Computer Science or a related field, or equivalent experience of a kind and amount that provides a comparable background
• Git and GitHub familiarity and hygiene, including pull requests, code review, and GitHub Actions CI
• 18+ months of daily use of AI coding agents and harnesses (e.g., Claude Code, Codex, Pi, OpenCode), with familiarity with agent loops and software factory pipelines, and work or a portfolio you can demonstrate
• Ability to read AI research and white papers with high comprehension
• Experience with retrieval-augmented generation (RAG) or graph RAG (e.g., Docling for ingestion, pgvector for retrieval, or LangChain-style equivalents)
• Experience building production web applications with TypeScript and React (1+ years, or equivalent projects)
• Dependency management best practices (e.g., npm or pnpm, uv or pip, Cargo; lockfiles, pinned versions and regular updates)
• Strong skills in Python, Node.js (TypeScript) and Rust, with REST API design experience (e.g., FastAPI)
• Working experience with PostgreSQL and SQLite, including schema migrations (e.g., SQLAlchemy, Alembic)
• Docker or Podman basics
• Test-first development (e.g., pytest, Vitest)
• Data hygiene, including safe handling of secrets and data
• Ability to work across frontend and backend within the same feature and to communicate trade-offs clearly
• Ability to take UX/UI designs and implement them faithfully
• Experience working in an Agile team

Professional Competencies

• Apply excellent analytical and problem-solving skills
• Communicate technical concepts clearly, in writing and verbally, to technical and non-technical audiences
• Work and interact effectively and professionally with and for others throughout various levels of the global organization
• Work effectively in both collaborative and independent environments with minimal supervision
• Remain calm and receptive in fast-paced situations
• Fluency in Artificial Intelligence (AI) and data concepts, including the ability to critically evaluate AI outputs, improve workflows using approved tools, identify and mitigate bias or inaccuracies, and communicate AI-driven insights clearly and responsibly to peers and managers

Preferred

• Experience developing interactive, data-driven interfaces, including dashboards, knowledge maps, and graph or network visualizations using tools such as D3.js
• Playwright and accessibility testing (WCAG); responsive interfaces for mobile and tablet devices
• Experience integrating real-time media and streaming technologies, such as LiveKit, WebRTC, or WebSockets, into voice or conversational applications
• Familiarity with handling personally identifiable information (e.g., FERPA or HIPAA; education or healthcare data)
• Kubernetes and Helm, including self-hosted clusters (e.g., OpenStack Magnum) alongside managed cloud services
• OIDC/SSO and Microsoft Entra ID (MSAL) in multi-role applications
• Experience working in startups, EdTech, healthcare technology, or other regulated environments that require close collaboration with institutional partners
• LLM-powered conversational products, including prompt design, multi-agent orchestration, evaluation of response quality, and basic statistics
• Self-hosted or open-weight models and OpenAI-compatible model serving
• Go as an additional backend language
• Mobile or desktop app development (Android, iOS, macOS, Electron)
• Operating services on Azure; Terraform or other infrastructure as code
• Familiarity with fine-tuning models
• Awareness of calibrated decision models
• Background workers and pipelines, including race conditions and idempotency
• Self-hosted Git and CI (e.g., Forgejo)
• Container/image security scanning (e.g., Trivy), semantic-version release workflows, and observability (e.g., OpenTelemetry)