DeepFlow is a London-based startup on a mission to build semi-autonomous companies through solving human-agentic orchestration. DeepFlow’s platform leverages advanced AI and multi-agent LLM technologies across five critical operational areas:
• Task Definition and Decomposition: Precisely breaking down complex goals into manageable sub-tasks.
• Task Delegation and Optimisation: Intelligently assigning tasks to the most suitable human or AI agents.
• Task Execution and Automation: Seamlessly carrying out defined tasks, leveraging automated processes where possible.
• Management and Coordination: Overseeing ongoing workflows and ensuring synchronised efforts across diverse agents.
• Governance, Security, and Compliance: Maintaining oversight and adherence to established protocols and regulations.
By embedding intelligent agents and agentic systems throughout these five areas, DeepFlow positions itself at the forefront of workflow management innovation.
Our ambitious research vision at DeepFlow is centred on building autonomous agentic systems capable of collaborating in dynamic multi-agent/human teams, transforming complex coordination into a significant human advantage. This endeavour involves cutting-edge research and innovative integration across several highly active fields of AI and machine learning, including high-capacity multi-modal foundation models, reinforcement learning, and multi-agent collaboration frameworks.
We are looking for a highly motivated intern to join our ML team led by Dr. Stefano Albrecht (Director of AI at DeepFlow). You will work with our ML engineers to design, implement, and evaluate agentic systems and multi-agent collaboration frameworks for workflow automation within the DeepFlow platform. We are aiming to generate research papers to submit to top-tier international ML conferences (e.g. NeurIPS, ICML, ICLR), and your work will directly contribute to our research and you will be involved in the publications processes.
Essential skills:
• Bachelor’s or Master’s degree (or higher) in computer science, AI/ML, or related field
• Proficiency with Python programming (e.g. OOP, Pydantic, FastAPI) and version control (git)
• Basic knowledge of LLMs and how they work
• Ability to work independently and in a team
• Clear communication at meetings
Desired skills:
• Experience in building/using AI/ML methods (e.g. LLMs, multi-agent systems, reinforcement learning, agents SDKs)
• Knowledge of agentic systems, reasoning models, tool use, etc.
• Experience contributing to empirical AI research projects (e.g. demonstrated via publications at top AI/ML conferences)
Internship details:
Start: August 2025 (flexible start date)
Duration: 6 months
Salary: £3,000/month
Location: Mindspace Shoreditch, London & Hybrid (wfh)
Benefits: we provide MacBook Pro, weekly team lunch
To apply to this role, please send the following materials via email to Stefano Albrecht - stefano.albrecht@deepflow.com - with subject line
“DeepFlow Internship”:
• CV
• 1-page letter describing your motivation for joining this role and how you match the essential/desired skills
• Two reference letters (or contacts for two referees who agreed to provide letters)
We will start to screen applications and interview shortlisted candidates after 21 July 2025.
Applications will be accepted until the position is filled.
Note: you must have the right to work full time in the UK. We cannot offer visa sponsorship at this stage.
For any questions about this role, please contact Stefano Albrecht.