Skip to content
← Lovable

AI Research Engineer, Post-Training

Lovable · London, United Kingdom

Applied AIonsite
Apply on the lab’s site ↗

At a glance

Highlights

  • Own the full post-training pipeline from data curation to deployment.
  • Ship improved code-generation and agent models to millions fast.
  • Move promising research into production within days or weeks.

Ideal candidate

A hands-on builder who has trained models at scale and lives to ship. You take extreme ownership, move fast from paper to production, and measure success by real user impact.

Requirements

  • Personally run post-training jobs on LLMs such as RFT/RLVR or preference optimization
  • Write solid production code for reliable systems
  • Fluent in PyTorch or JAX with distributed training and GPU clusters experience
  • Understand math behind preference optimization, reward modeling and alignment
  • Built evaluation systems capturing real-world quality beyond benchmarks
  • Trace model quality regressions from user symptoms through serving to training
  • Driven to ship research to users quickly

Lab profile

About Lovable

AI platform that turns plain-language prompts into production-ready software.

What they build

  • AI software creation platform
  • Plain-language to full-stack app builder

Known for

  • LLM post-training for code generation and agents
  • Applied AI systems and agents for operations
  • AI-native brand and web production
  • Agent-first data and analytics platform

Key facts

  • 2M+ people building software on platform
More about Lovable →
Apply on the lab’s site ↗

First seen · Last seen