Founding AI Researcher

TL;DR

Founding AI Researcher (LLMs): Building an end-to-end code generation and boilerplate elimination system that turns English descriptions into correct code with real-time feedback. Focus on / с акцентом на LLM serving and fine-tuning, agentic/multi-agent evaluation, and inference optimization to reduce latency and cost.

Location: On-site, London, UK

Company

hirify.global builds a higher-level programming language that preserves human intent while using LLMs to generate and refine code.

What you will do

• Develop a code generation pipeline that turns English descriptions/specs into code.
• Build a boilerplate elimination pipeline that converts trivial code into concise English descriptions.
• Create evaluations for both pipelines to ensure generated code correctness and equivalence.
• Improve code generation stability and correctness, including equivalence to replaced boilerplate.
• Implement real-time feedback (as users type) to support editing English descriptions in an IDE-like workflow.
• Design a code merge pipeline that merges English intent while generating and merging the corresponding code, plus latency improvements.

Requirements

• Strong applied ML foundation, from classic ML to modern neural networks.
• Expertise in LLMs, including serving, fine-tuning (SFT, LoRA), and quality evaluation.
• Experience designing and evaluating agentic/multi-agent systems from 0 to 1.
• Strong inference optimization and agentic workflow tuning skills (quantization, batching, KV-cache, context engineering) to reduce latency and cost.
• On-site in London, UK.

Culture & Benefits

• Startup mode with ownership, exploration, and cross-discipline collaboration.
• Work alongside engineers and AI researchers from companies including Kotlin/JetBrains, Databricks, Isomorphic Labs, and Google.
• Focus on delivering results that make users happy.
• Encouragement of cross-discipline work and practical impact.

Hiring process

• Not specified in the provided posting.