Research Engineer – AI Agents & LLM Systems

Here is a client-neutral version with identifying company references removed and the wording made more market-facing.

Research Engineer – AI Agents & LLM Systems

About the Role

We are seeking a highly capable Research Engineer to join an advanced AI Agent research team.

You will research, build, and evaluate next-generation AI agent systems that use large language models and multimodal foundation models to reason, plan, interact with tools, and complete complex multi-step tasks.

This is a hands-on research engineering role suited to someone who enjoys turning open-ended technical questions into working systems. You will combine strong software engineering with sound research judgment, taking ownership of problems from initial investigation through implementation, evaluation, failure analysis, and iterative improvement.

Key Responsibilities

• Design and implement AI agents based on large language models and multimodal foundation models.

• Develop agent systems incorporating tool use, planning, memory, retrieval, code execution, computer interaction, and multi-step workflows.

• Build rigorous evaluation environments, benchmarks, and failure-analysis tooling for agent capabilities.

• Prototype new research ideas rapidly and develop successful approaches into robust, reusable implementations.

• Reproduce, evaluate, and challenge methods from recent AI and machine learning research.

• Design and run controlled experiments to support evidence-based technical decisions.

• Improve agent reliability, task-completion performance, latency, computational efficiency, and operating cost.

• Develop clean and reusable research infrastructure, APIs, tooling, and technical demonstrations.

• Investigate failure modes across reasoning, planning, retrieval, memory, and tool execution.

• Collaborate with researchers and engineers while independently owning substantial technical projects.

• Communicate research findings clearly through code, technical documentation, reports, presentations, and demonstrations.

Essential Requirements

• MSc or PhD in Computer Science, Artificial Intelligence, Machine Learning, or a related technical discipline, or equivalent practical experience.

• Strong Python programming and software engineering skills.

• Experience designing and implementing non-trivial machine learning or artificial intelligence systems.

• Practical understanding of large language models, transformer architectures, and modern generative AI systems.

• Hands-on experience with PyTorch or an equivalent deep learning framework.

• Ability to design meaningful experiments, benchmarks, and evaluation methodologies.

• Strong debugging, analytical, and problem-solving skills.

• Evidence of independently delivering substantial technical projects.

• Ability to work effectively in an ambiguous and rapidly evolving research environment.

• Strong written and verbal communication skills in English.

Knowledge of TypeScript is advantageous but not required.

Preferred Qualifications

• Two or more years of relevant industry, research engineering, or applied research experience.

• Experience developing tool-using AI agents, coding agents, computer-use systems, autonomous workflows, or other multi-step LLM applications.

• Experience with agent orchestration, planning, memory systems, retrieval-augmented generation, and structured tool calling.

• Experience with LLM evaluation, agent trajectory analysis, or automated evaluation systems.

• Knowledge of reinforcement learning, model post-training, preference optimisation, or synthetic data generation.

• Experience deploying, serving, or optimising open-source language models.

• Familiarity with inference optimisation, model serving, distributed systems, or GPU-based AI infrastructure.

• Contributions to open-source AI, machine learning, or agent-development projects.

• Publications at recognised AI, machine learning, NLP, or systems conferences.

• Experience progressing an AI system from research prototype through to production or real-world users.

Candidate Profile

The successful candidate is likely to combine research curiosity with strong engineering execution. Academic publications are valued but are not essential.

Greater emphasis will be placed on:

• Technical depth and engineering quality.

• Independent thinking and problem ownership.

• Ability to build working AI systems rather than only conceptual prototypes.

• Strong experimental methodology and objective evaluation.

• Ability to identify and resolve complex failure modes.

• Experience translating research ideas into reliable implementations.

• Ability to operate effectively where the correct technical approach is not known in advance.