AI-First Full-Stack Engineer — End-to-End LLM Apps, Equity

bifrost; Barcelona; Presencial; EUR 70.000 - 100.000; Jornada completa

Descripción de la vacante

Shortlists is an AI-native recruitment CRM and ATS. This role focuses on AI engineering as much as full-stack work, owning surfaces end-to-end from UI to backend deployment and edge logic. You will build and operate LLM-backed features, turn customer feedback into shipped product, and work with agents as a core workflow.

Strong TS/React, CI/CD discipline, and real code shipping are required. The team works fast with async-first communication and regular in-person time.

Formación

• Shipped full-stack features with migrations, auth, deploys and 2am production debugging.
• Built or integrated with an LLM in the critical path and kept it working.
• Experience in a real engineering pipeline with branches, CI, reviews and releases.
• You can scope problems, review agent outputs, and keep judgment on architecture, data and security.
• You can read and reason about code you did not write and catch plausible but wrong outputs.

Responsabilidades

• Own one or two product surfaces end to end: UI, database schema, edge functions, deployment and post-launch behavior.
• Build and operate the LLM-backed parts of the product: prompts, retrieval, evals, cost and latency management.
• Turn customer objections or transcripts into shipped product with informed judgment.
• Work with agents as a standard workflow and oversee their outputs.
• Improve the codebase with sharper rules and gating to catch issues before PRs.
• Communicate with customers to understand and fix what is broken.

Conocimientos

TypeScript

React

CI/CD

Code review

Agent-based AI

Postgres

Herramientas

Claude Code

Codex

Postgres

Supabase

Next.js

Vercel

Descripción del empleo

Shortlists is an AI-native recruitment CRM and ATS. This role focuses on AI engineering as much as full-stack work, owning surfaces end-to-end from UI to backend deployment and edge logic. You will build and operate LLM-backed features, turn customer feedback into shipped product, and work with agents as a core workflow.

Strong TS/React, CI/CD discipline, and real code shipping are required. The team works fast with async-first communication and regular in-person time.