Junior Full-Stack Developer

ELYTS LLC; United States; Remote; USD 90,000 - 120,000; Full time

Job summary

ELYTS LLC is seeking a Junior Full-Stack Engineer to build and ship features with generative AI directly in users' hands. You will work across the stack—frontend chat interfaces, backend LLM orchestration, and data plumbing—collaborating with senior engineers, designers, and product teams to turn AI capabilities into reliable user-facing functionality.

This hands-on role is ideal for someone early in their career who is curious about LLM systems and eager to gain production experience working

Qualifications

• Practical experience building web applications end-to-end.
• Solid foundation in at least one modern frontend framework (React/Next.js/Vue/Svelte).
• Backend experience in Node.js, Python, or Go with REST/GraphQL APIs.
• Hands-on use of at least one LLM API (OpenAI/Anthropic/Google or open-source).
• Understanding GenAI concepts: prompts, system prompts, tokens, context windows.
• Familiar with Git, branching workflows, and CI/CD practices.
• Strong written communication for explaining technical trade-offs.

Responsibilities

• Build and maintain user-facing AI features such as chat interfaces, document Q&A, summarization tools, and inline assistants.
• Develop frontend components that handle streaming LLM responses, loading states, retries, and graceful error handling.
• Implement backend endpoints and services that orchestrate LLM calls, manage conversation context, and apply business logic around model outputs.
• Write and refine prompts, system instructions, and tool/function definitions for production use cases.
• Integrate embeddings and vector retrieval (e.g., pgvector, Pinecone, Weaviate) for retrieval-augmented generation over internal content.
• Add logging, tracing, and basic evaluation hooks so the team can monitor quality, latency, and cost.
• Write tests — including for non-deterministic AI outputs — and participate in code review.
• Contribute to technical decisions as you grow, and help document patterns the rest of the team can reuse.

Skills

React

Next.js

Vue

Svelte

Node.js

Python

Go

LLM APIs

GenAI concepts

Git

CI/CD

TypeScript

Tools

pgvector

Pinecone

Weaviate

WebSockets

Server-Sent Events

Job description

About the Role

We are looking for a Junior Full-Stack Engineer to help build and ship product features that put generative AI directly in users' hands. You will work across the stack — frontend chat and assistant interfaces, backend services that orchestrate LLM calls, and the data plumbing around them — collaborating closely with senior engineers, designers, and product to turn AI capabilities into reliable, user-facing functionality. This is a hands‑on role for someone early in their career who is genuinely curious about LLM systems and wants to build real production experience with them.

Responsibilities

• Build and maintain user‑facing AI features such as chat interfaces, document Q&A, summarization tools, and inline assistants.


• Develop frontend components that handle streaming LLM responses, loading states, retries, and graceful error handling.


• Implement backend endpoints and services that orchestrate LLM calls, manage conversation context, and apply business logic around model outputs.


• Write and refine prompts, system instructions, and tool/function definitions for production use cases.


• Integrate embeddings and vector retrieval (e.g., pgvector, Pinecone, Weaviate) for retrieval‑augmented generation (RAG) over internal content.


• Add logging, tracing, and basic evaluation hooks so the team can monitor quality, latency, and cost.


• Write tests — including for non‑deterministic AI outputs — and participate in code review.


• Contribute to technical decisions as you grow, and help document patterns the rest of the team can reuse.

Requirements

• Practical experience (internships, bootcamp projects, contributions, or 1+ year professional) building web applications end-to-end.


• Solid foundation in at least one modern frontend framework (React, Next.js, Vue, or Svelte) including component composition, state management, and async data fetching.


• Backend experience in Node.js, Python, or Go: building REST or GraphQL APIs, handling I/O, working with JSON.


• Hands‑on use of at least one LLM API (OpenAI, Anthropic, Google, or open‑source via Ollama/vLLM/Hugging Face).


• Understanding of core GenAI concepts: prompts vs. system prompts, context windows, tokens, temperature, and the difference between chat and completion APIs.


• Comfort with Git, branching workflows, and basic CI/CD.


• Strong written communication; ability to explain technical trade‑offs clearly.

Nice to Have

• Experience with streaming responses (Server‑Sent Events, WebSockets, fetch streams).


• Familiarity with vector databases, embeddings, and retrieval pipelines.


• TypeScript across the stack.


• Experience deploying to a cloud platform (AWS, GCP, Vercel, Fly.io, etc.).


• Observability tooling (LangSmith, Langfuse, OpenTelemetry, Datadog).


• Side projects, open‑source contributions, or writing about AI you have shipped or studied.