About the role
Responsibilities Build internal AI services using open-weight language models for document processing and support Connect models to company systems through APIs with validation, access controls and fallback Deploy models on company-controlled infrastructure and improve latency, capacity and serving cost Run repeatable quality, latency and cost evaluations comparing internal models with external providers Integrate models with coding assistants and agents within approved permissions Track usage, operating costs and time saved after review Requirements 3+ years in software engineering, ML engineering or applied AI Experience delivering an LLM application used by real users Strong Python skills, including APIs and data processing Hands-on experience in LLM application development, model serving, evaluation or fine-tuning, or coding agents Ability to design repeatable evaluations and explain trade-offs English for documenting results with international teams Preferred Qualifications Experience with vLLM or SGLang, load testing and GPU memory optimisation Experience fine-tuning models on domain data with held-out validation Experience with document-grounded retrieval or coding agents integrated with repositories Benefits Remote-first company Competitive salary with quarterly performance bonuses 28 days paid annual leave Referral and flash bonuses Top of the line equipment Annual company retreats About the Company Sporty is a remote-first company in pursuit of sustainability.