Epoch Biodesign
Machine Learning Engineer Epoch Biodesign is a well funded, venture backed start up using biology to make every type of plastic recyclable - starting with nylon. Using a unique combination of AI, synthetic biology and green chemistry, we are scaling enzymatic recycling in order to transform currently unrecyclable plastics and textiles into new, virgin quality materials. Our technology yields substantial reductions in carbon emissions with disruptive unit economics, preventing waste from entering landfill or the environment, allowing us to solve this very urgent challenge. With our pilot plant already processing nylon 6,6 waste, we will imminently complete construction on our larger demo facility. This site will produce material destined for use in garments made by some of the world's biggest fashion, sportswear and luxury brands, and also in components for some of the world's largest car companies. The Role As a Machine Learning Engineer at Epoch you will design and build custom machine learning models that deepen our understanding of protein sequence-structure-function relationships, directly informing enzyme engineering decisions. Working closely with computational and lab biologists, you will rapidly prototype approaches, figure out what works and what doesn't, and develop custom models - not off the shelf solutions - grounded in evolutionary constraints. We want someone scrappy and pragmatic who can move fast, test ideas cheaply, and pivot without hesitation. Your core activities will include: Building custom machine learning and probabilistic models to capture meaningful patterns in protein sequence and structure data, with an emphasis on understanding evolutionary and functional relationships Developing hierarchical and Bayesian modelling approaches to integrate heterogeneous experimental data (e.g. high throughput screening results alongside detailed biochemical assays) Rapidly prototyping against real datasets, quickly assessing whether an approach has merit, iterating based on experimental feedback, and knowing when to abandon an idea and try something different Acting as primary contact with computational and lab biologists to understand their modelling requirements and translate biological questions into well framed ML problems Communicating model outputs and their implications clearly to both wet and dry lab colleagues and to management Training colleagues in the use of deployed models and methods Transitioning successful prototypes into maintainable production tools Essential Qualifications and Experience A degree (PhD preferred) in a quantitative or life science field such as computational biology, bioinformatics, mathematics, computer science, physics, biochemistry or similar - with demonstrated application to biological problems Hands on experience applying machine learning in a life science or biotechnology context Demonstrated ability to build custom models from scratch, not solely relying on pre trained or foundation models - including experience with probabilistic and/or Bayesian modelling approaches Strong technical background in Python and common scientific/ML libraries (e.g. NumPy, SciPy, Pandas, scikit learn, PyTorch or equivalent) Experience working with protein sequence and/or structure data Experience with environment and dependency management tools (e.g. Conda, Mamba) Skills in data engineering, cleaning and preprocessing of experimental datasets Skills in data visualisation (e.g. Matplotlib, Plotly or equivalent) Excellent interpersonal skills, with demonstrable evidence of working collaboratively in an interdisciplinary environment Experience with version control systems (Git) Beneficial Experience with deep learning methods applied to biological data Experience with large language models (LLMs) or protein language models Demonstrated experience with Google Cloud Platform (GCP) or Amazon Web Services (AWS), including cloud based pipeline orchestration Proficiency with Docker and containerised workflows Skills in automated testing Experience in project management Previous start up or scale up experience A working understanding of ClaudeCode or similar Benefits and perks Epoch Biodesign offers a comprehensive benefits program. At the moment this includes: A generous allowance of 30 days paid holiday (plus the usual 8 bank holidays) Meaningful EMI Share Options A non contributory pension of 9% employer contribution Optional company covered private medical insurance with Vitality Group Income Protection Group Critical Illness Flexible working around the core times of 10am to 4pm Cycle to work scheme The opportunity to be part of building something remarkable On-the-job perks Complementary fresh fruit, coffee, tea and snacks Onsite gym Various staff social activities And we're continuously reviewing and enhancing our benefits and work environment.
Machine Learning Engineer Epoch Biodesign is a well funded, venture backed start up using biology to make every type of plastic recyclable - starting with nylon. Using a unique combination of AI, synthetic biology and green chemistry, we are scaling enzymatic recycling in order to transform currently unrecyclable plastics and textiles into new, virgin quality materials. Our technology yields substantial reductions in carbon emissions with disruptive unit economics, preventing waste from entering landfill or the environment, allowing us to solve this very urgent challenge. With our pilot plant already processing nylon 6,6 waste, we will imminently complete construction on our larger demo facility. This site will produce material destined for use in garments made by some of the world's biggest fashion, sportswear and luxury brands, and also in components for some of the world's largest car companies. The Role As a Machine Learning Engineer at Epoch you will design and build custom machine learning models that deepen our understanding of protein sequence-structure-function relationships, directly informing enzyme engineering decisions. Working closely with computational and lab biologists, you will rapidly prototype approaches, figure out what works and what doesn't, and develop custom models - not off the shelf solutions - grounded in evolutionary constraints. We want someone scrappy and pragmatic who can move fast, test ideas cheaply, and pivot without hesitation. Your core activities will include: Building custom machine learning and probabilistic models to capture meaningful patterns in protein sequence and structure data, with an emphasis on understanding evolutionary and functional relationships Developing hierarchical and Bayesian modelling approaches to integrate heterogeneous experimental data (e.g. high throughput screening results alongside detailed biochemical assays) Rapidly prototyping against real datasets, quickly assessing whether an approach has merit, iterating based on experimental feedback, and knowing when to abandon an idea and try something different Acting as primary contact with computational and lab biologists to understand their modelling requirements and translate biological questions into well framed ML problems Communicating model outputs and their implications clearly to both wet and dry lab colleagues and to management Training colleagues in the use of deployed models and methods Transitioning successful prototypes into maintainable production tools Essential Qualifications and Experience A degree (PhD preferred) in a quantitative or life science field such as computational biology, bioinformatics, mathematics, computer science, physics, biochemistry or similar - with demonstrated application to biological problems Hands on experience applying machine learning in a life science or biotechnology context Demonstrated ability to build custom models from scratch, not solely relying on pre trained or foundation models - including experience with probabilistic and/or Bayesian modelling approaches Strong technical background in Python and common scientific/ML libraries (e.g. NumPy, SciPy, Pandas, scikit learn, PyTorch or equivalent) Experience working with protein sequence and/or structure data Experience with environment and dependency management tools (e.g. Conda, Mamba) Skills in data engineering, cleaning and preprocessing of experimental datasets Skills in data visualisation (e.g. Matplotlib, Plotly or equivalent) Excellent interpersonal skills, with demonstrable evidence of working collaboratively in an interdisciplinary environment Experience with version control systems (Git) Beneficial Experience with deep learning methods applied to biological data Experience with large language models (LLMs) or protein language models Demonstrated experience with Google Cloud Platform (GCP) or Amazon Web Services (AWS), including cloud based pipeline orchestration Proficiency with Docker and containerised workflows Skills in automated testing Experience in project management Previous start up or scale up experience A working understanding of ClaudeCode or similar Benefits and perks Epoch Biodesign offers a comprehensive benefits program. At the moment this includes: A generous allowance of 30 days paid holiday (plus the usual 8 bank holidays) Meaningful EMI Share Options A non contributory pension of 9% employer contribution Optional company covered private medical insurance with Vitality Group Income Protection Group Critical Illness Flexible working around the core times of 10am to 4pm Cycle to work scheme The opportunity to be part of building something remarkable On-the-job perks Complementary fresh fruit, coffee, tea and snacks Onsite gym Various staff social activities And we're continuously reviewing and enhancing our benefits and work environment.