AI Engineer: Post-Training (AI)

Location: London, United Kingdom

Company

hirify.global is a physics AI company building simulation software for industrial engineering and manufacturing across aerospace and defence, automotive, semiconductors, materials, and energy.

What you will do

• Own the path from base models and customer workflows to measured, deployable model improvements.
• Build post-training harnesses covering data preparation, training, evaluation, model artifacts, and hand-off to serving.
• Apply SFT, preference optimisation, RLHF/RLAIF, verifiable-reward reinforcement learning, distillation, rejection sampling, and related methods.
• Build synthetic- and expert-data pipelines with quality filtering, provenance, licensing, and contamination controls.
• Design reward signals, verifiers, and rollout workflows while detecting reward hacking, benchmark overfitting, model collapse, and regressions.
• Set the capability roadmap and deliver managed workflows and expert-facing interfaces across SaaS, VPC, and air-gapped customer environments.

Requirements

• Measurably improved model behaviour through post-training and turned the work into a production capability.
• Hands-on experience across data, training, evaluation, debugging, and taking experiments from hypothesis to deployed artifact.
• Strong understanding of modern post-training methods and the ability to select methods based on task requirements and evidence.
• Experience building high-quality expert, behavioural, or synthetic training datasets; understanding of data design as part of model design.
• Ability to diagnose instability, drift, reward exploitation, leakage, and evaluation overfitting.
• Product-minded approach to defining user-workflow success, making trade-offs, and building operable technical capabilities.
• Strong Python and PyTorch or equivalent model-training frameworks.
• Experience with distributed training, accelerator utilisation, checkpointing, recovery, experiment tracking, reproducibility, model evaluation, controlled experiments, and ablations.
• Knowledge of data governance, privacy, provenance, customer-owned training data and artifacts, model packaging, and production serving.
• Clear technical writing and communication across research, product, engineering, delivery, and customer teams.

Nice to have

• Post-training open-weight models for tool use, reasoning, or long-horizon agent workflows.
• Large-scale rollout generation, reward models, verifiers, or online reinforcement learning.
• Managed and expert-facing model customisation products.
• Contributions to post-training research or open-source training frameworks.
• Experience with engineering, simulation, design, manufacturing, regulated, customer-controlled, or air-gapped environments.

Culture & Benefits

• Work with Research, platform engineering, and product teams on industrial physics AI capabilities.
• Customer data and model artifacts are designed to remain within customer-controlled environments.
• Commitment to equal employment opportunity and diversity across sex, race, religion, ethnicity, nationality, disability, age, sexual orientation, and gender identity.
• Diversity and inclusion data is collected confidentially and used only in aggregate to monitor equal-opportunity policies.