Software Engineer II -AI Platform {Teacher Assistant)

Remote

Impact the Moment

At client , our AI Platform team is building intelligent learning experiences used by millions of students and educators. Teacher Assistant is a direct-to-teacher AI tool - an agentic chatbot, embedded in the learning platform teachers already use, that helps K-12 teachers plan instruction, find and use their course materials, and make sense of student performance data through natural conversation. This is applied AI with real stakes. T he orchestration is genuinely hard, and the impact - helping a teacher reach a student who//'s been struggling - is something you//'ll actually feel.

About this engagement

You//'ll join a small, senior delivery team building and operating the Teacher Assistant backend and the AI platform around it. We//'re looking for someone who can pick up well-scoped features and own them through to production with light support - someone who has shipped production AI systems before and understands how they behave once real users are in the loop. A note on shape: this is a full-stack-leaning-backend role with a strong AI-systems emphasis. It is not a pure machine-learning role and not a pure web role. Most of your time is in async Python and the LLM orchestration layer; you//'ll touch the frontend when the work calls for it.

What you//'ll do

Day to day, you//'ll:

• Build and extend the agentic LLM orchestration behind Teacher Assistant - the graph of nodes and agents that turns a teacher//'s request into a useful

response .

• Integrate external data sources and tools into the agent so it can reason over the information teachers need .

• Improve retrieval quality across vector and lexical search .

• Work on model routing, resilience, and graceful degradation so the system stays fast and available under real-world load.

• Strengthen the prompt lifecycle, evaluation discipline, and observability that keep a nondeterministic system reliable .

• Harden quality with automated unit, integration, and end-to-end testing

wnat tn1s ro1e 1s - ana 1sn·t

This is a software engineering role on a production AI system. You won//'t be training or fine-tuning models or running model-science experiments - our data science team owns that. But unlike a generic application role, prompt engineering, retrieval quality, and evaluation discipline are core to this job, not someone else//'s problem. The interesting work lives in the orchestration, the prompt lifecycle, retrieval, and evals - not CRUD.

What the role looks like at this level

As a Software Engineer II on this team, you//'ll break down medium-sized features, estimate them, and cut scope to ship on time. You//'ll start to own tasks within the service with support from senior teammates, contribute to technical design and engineering-review proposals while thinking through failure cases, give helpful and timely code reviews, and defend your decisions in review. You//'ll debug to root cause in your area, instrument your code for operations, and participate in the on-call rotation. Senior engineers are around to pair with and review your work - but increasingly you//'ll be the one proposing the approach and carrying a feature to production.

What you must already bring

You don//'t need every line below at expert depth, but the combined surface has to be covered.

Core engineering

. Expert-level async Python (3.11+). Real production asyncio / async / await experience across the request path - a synchronous-only Python background won//'t be enough here.

• FastAPI at depth: routers, dependencies, lifespan, middleware. Pydantic v2 and disciplined type hints.
• pytest and pytest-asyncio - fixtures, async, mocking, and meaningful coverage. Standard formatting, linting, and type-checking tools are table stakes.

AI / LLM systems - the heart of the work

• Hands-on production experience with LangGraph: state machines, conditional edges, checkpointing. Experience with LangChain alone is not the same thing - this is where most of the surface area lives.
• LangChain core (messages, runnables, tools), and prompt engineering/ prompt lifecycle management - versioned, environment-tagged prompts with local overrides - using tracing and experiment tooling such as LangSmith .
• Multi-agent/ multi-node workflow design - routing across specialized agents and nodes.

• RAG with hybrid vector + lexical retrieval, and experience with a managed LLM provider such as Azure OpenAI (deployments, API versions, quotas).
• Sound instincts for non-determinism, token budgets, timeouts, and graceful degradation,

plus familiarity with eval frameworks (e.g. LLM-as-judge and regression evals).

Data, infrastructure, and delivery

. PostgreSQL operationally - indexing, connection pools, poolers - plus pgvector and

OpenSearch/Elasticsearch hybrid (text + KNN) search.

. AWS and Kubernetes in production - genuine fluency, beyond local container orchestration. Docker multi-stage builds; infrastructure-as-code (e.g. Terraform) and manifest overlays for multiple environments.

. Multi-environment configuration discipline - several environments, from local through production, each with its own secrets, prompts, and resources.

And comfortable with

• Typescript and modern Angular with RxJS when frontend work is needed. A backend-leaning candidate is welcome as long as you//'re comfortable in Angular; a frontend-leaning candidate must still be solid in the Python/LLM stack.

Nice to have {genuine bonuses, none required)

• MCP (Model Context Protocol) and SSE; database migration tooling; Redis-compatible caches.
• Observability tooling (APM, metrics, tracing) and distributed-tracing concepts.
• Modern Python packaging and build tooling, Make-based builds, GitHub Actions, private package registries, and encrypted-secrets workflows.
• Load testing and end-to-end browser testing frameworks.
• Edtech / K-12 domain awareness (standards, proficiency, learning frameworks) and FERPA-adjacent data-privacy thinking.
• Familiarity with large-enterprise internal identity, auth, and content-metadata services - accelerates ramp, but learnable.

How we work

This is an internal enterprise codebase, so expect internal SDKs and package registries, encrypted-secrets tooling, and a secrets manager as part of the daily flow. It//'s a polyglot repo - backend, frontend, infrastructure-as-code, and database migrations coexist - and the team uses written design and decision docs. Security hygiene for AI apps (prompt injection, PII handling, guardrails) matters here because we//'re working with educational data.