Postdoctoral scholarship (2 years) within the Trustworthy Al for wastewater treatment through formal physical and chemical constraints project.

The Department of computing science is offering a postdoctoral scholarship within the project” Trustworthy Al for wastewater treatment through formal physical and chemical constraints”. The scholarship is full-time for two years with access November 1, 2026 or by agreement. 

The Department of Computing Science at Umeå University is an international environment for research and education covering a broad spectrum of computer science. The department is internationally renowned for its excellent research and education. We are affiliated with WASP – Sweden’s largest individual research program of all time, coordinate its sister program WASP-HS, which focuses on the consequences of emerging technologies, and are a strong partner in the national strategic research program eSSENCE. We value diversity and actively work to promote an inclusive work environment with a high level of engagement. Read more at: Department of Computing Science 

The Formal Methods for Trustworthy Hybrid Intelligence research group at Umeå University focuses on designing intelligent systems that provide formal behavioral guarantees, with the goal of empowering human collaborators rather than replacing them. Their extensive research interests encompass neuro-symbolic AI, logic-based learning, formal argumentation, common-sense reasoning, AI ethics and testing, and cognitive modeling. The group applies these theoretical foundations to a diverse range of practical projects, such as developing trustworthy decision support systems for energy management, utilizing mixed reality for situational awareness in police interventions, creating smart personalized medication assistants, and researching precision medicine for injury recovery, while also contributing to AI education and strategic interaction models. 

Project description
This two-year postdoctoral project offers a research grant funded by the Kempestiftelserna (Stiftelserna J.C. Kempes och Seth M. Kempes minne) , hosted at the Department of Computing Science at Umeå University.  The scholarship is tax-free and amounts to 750,000 SEK for two years, corresponding to 375,000 SEK per year. 

Modern Wastewater Treatment Plants (WWTPs) increasingly utilize AI-based decision support systems to optimize purification processes. However, current data-driven artificial intelligence approaches operate purely on statistical correlations, meaning they can generate recommendations that violate fundamental chemical and physical laws. This lack of physical grounding severely limits their reliability and adoption within safety-critical environmental infrastructure.   

The primary objective of this research initiative is to establish a scientifically grounded neurosymbolic AI framework for wastewater treatment. By integrating symbolic reasoning mechanisms with neural networks trained on real-world operational data, the project aims to ensure that AI-guided decisions remain physically consistent, interpretable, and aligned with immutable laws (e.g., physical and chemistry equations). The project is conducted in close academic collaboration with regional industrial actors, providing a unique real-world testbed for verifying these advanced AI models against strict European and national environmental regulations.   

Research Opportunities and Areas of Exploration
As a scholarship recipient, you will be provided with the opportunity to independently deepen and develop your scientific profile within the field of neurosymbolic AI. The scholarship offers a stimulating research environment in which you can build upon your knowledge and academic merits through the exploration of, among others, the following areas: 

1. Advancing knowledge at the forefront of neurosymbolic AI. Opportunity to deepen understanding of emerging neurosymbolic frameworks that integrate neural networks for time-series analysis (e.g., LSTM) with logic-based AI methods (e.g., Answer Set Programming) in applications related to industrial automation. 
2. Exploration of physical and chemical principles in AI models. Opportunity to investigate how physical and chemical laws relevant to wastewater treatment can be represented and incorporated into symbolic knowledge structures for advanced AI systems. 
3. Development of expertise in hybrid AI methodologies. Opportunity to gain deeper insight into concepts, methods, and experimental approaches for hybrid AI architectures that combine deep learning with symbolic knowledge representations. 
4. Advancing understanding of methods for analyzing and interpreting AI systems. Opportunity to explore approaches for studying the performance, safety, and interpretability of neurosymbolic models in industrial application contexts. 
5. Academic merit development and international scientific engagement. Opportunity to strengthen an international research profile through scientific exchange, participation in international conferences, and publication of research findings in scientific journals. 
    

Qualifications
To qualify as a postdoctoral scholarship holder, the postdoctoral fellow is required to have completed a doctoral degree or a foreign degree deemed equivalent to a doctoral degree. This qualification requirement must be fulfilled no later than at the time of the decision about scholarship recipient. 

Priority should be given to candidates who completed their doctoral degree, according to what is stipulated in the paragraph above, no later than three years prior. If there are special reasons, candidates who completed their doctoral degree before that may also be eligible. Special reasons include absence due to illness, parental leave, appointments of trust in trade union organisations, military service or similar circumstances, as well as clinical practice or other forms of appointment/assignment relevant to the subject area. 

Doctoral degree should be within computing science or a closely related, relevant subject area with an artificial intelligence specialization.  

Formal requirements:

• A doctoral degree or a foreign degree (outside Sweden) deemed equivalent to a doctoral degree in computing science or a relevant subject area, completed no later than at the time of the scholarship decision. 
• Strong written and oral proficiency in English, as it is the primary language used for international scientific dissemination and collaboration within the research group. 

Requirements for the project:

• A solid academic background and documented research experience in artificial intelligence, with a specific focus on either symbolic AI (such as knowledge representation and reasoning, logic-based systems, or Answer Set Programming) or sub-symbolic AI (such as deep learning and neural network architectures). 
• Proficient software development skills (Python) to facilitate the independent implementation and simulation of novel hybrid AI prototypes. 

Other requirements:

• Experience working with or processing real-world data models originating from complex industrial settings or automated engineering processes. 
• A track record of publishing scientific results in peer-reviewed international conferences or high-impact journals within the field of computer science or AI. 

Meritorious qualifications:

• Prior experience in formalizing mathematical representations of physical or chemical laws into computational logic or formal constraints. 
• Familiarity with neurosymbolic frameworks such as ASP, NeurASP, DeepProbLog, or similar logic-neural interfaces. 
• Experience or familiarity with industrial automation domains, safety-critical environments, or environmental infrastructure systems like wastewater treatment plants. 

Personal qualities:

We are looking for you who possess a high degree of scientific curiosity, academic independence, and an analytical approach to complex theoretical problems. Since the project operates within a collaborative ecosystem involving both academic researchers and regional industrial actors, you should exhibit excellent communication and interpersonal skills. You are self-motivated, structured in your scientific methodology, and have a strong drive to explore interdisciplinary boundaries between data-driven computer science and physical domain knowledge 

Application
The application should be written in English or Swedish, and attached documents should be in Word or PDF format. The application should be registered via Umeå University’s e-recruitment system Varbi and submitted by the deadline 31 august, 2026. 

A full application should include: 

• Cover letter: A statement of interest (maximum 2 pages or 8,000 characters including spaces) detailing your scientific motivation for joining this project. The cover letter should outline:   
  - A brief research statement or proposal describing how you would approach the integration of formal physical/chemical constraints with neural networks within the scope of wastewater treatment.   
  - A description of how your previous academic achievements, research experience, and methodology relate to the core themes of this neurosymbolic AI project.   
  - An overview of your technical skills in AI development, knowledge representation, or logic-based systems that will enable you to independently pursue this research.   
• Curriculum vitae (CV) with publication list,
• Verified copy of doctoral degree certificate or documentation that clarifies when the degree of doctor is expected to be obtained,
• Verified copies of other diplomas, list of completed academic courses and grades,
• Copy of doctoral thesis and up to 2 articles written by the applicant.
• Other documents that the applicant wishes to claim.
• Contact information to two persons willing to act as references. 

Further information
For further information, please contact associate professor, Esteban Guerrero, esteban.guerrero@umu.se 

We look forward to receiving your application!