Senior Applied Scientist

Company Description

Truelogic AI Ltd is at the forefront of advancing artificial intelligence. We specialize in designing AI systems that integrate multilayered, scalable knowledge graphs with state-of-the-art reasoning agents. Our mission is to analyze and deeply understand intricate knowledge domains with unparalleled scholarly precision. As a growing innovative company, Truelogic AI fosters a collaborative environment to tackle challenging AI problems.

Role Description

We are looking for an Applied Scientist who will sit at the frontier where formal knowledge representation, large-scale generative AI, formal reasoning, theoretical CS/Optimization, knowledge graphs, and neuro-symbolic reasoning converge. You will design and deliver algorithms, methods and systems that are capable, scalable, correct, interpretable, and grounded in structured world knowledge — a combination that remains one of the open problems in applied AI. 

You will work in a multi-disciplinary team of SE, AI, and ML researchers, engineers, and content creators.  

This role demands rare depth: the ability to publish at the leading venues and the operational discipline to ship production systems used worldwide. You will define technical direction, mentor engineers, and operate as a technical voice in cross-functional strategy. 

What You Will Work On 

• Knowledge-Grounded Generation. Architect hybrid architectures that couple large language models (LLMs) foundation models with enterprise and open-domain knowledge graphs (KGs), ontologies, and symbolic rule engines to reduce hallucination and enforce logical consistency. 
• Neuro-Symbolic Reasoning Systems. Design and implement frameworks — drawing on approaches such as DeepProbLog, LNN (Logic Tensor Networks), LARK, or novel alternatives — that allow neural components to learn from structured logical constraints and symbolic components.  
• Formal Knowledge Representation. Develop automated tools that maintain, mine, refine and process ontologies, description logic schemas, and first-order logic formalisms (OWL 2, SPARQL, Datalog±) to power downstream reasoning modules and structured retrieval pipelines. 
• Retrieval-Augmented and Tool-Augmented AI. Build advanced RAG pipelines that leverage semantic, lexical, and structured (SPARQL/Cypher) retrieval over heterogeneous corpora with provenance tracking and citation grounding. 
• LLM Fine-Tuning and Alignment. Lead RLHF, DPO, or RLAIF experiments to align generative models to domain-specific constraints and safety requirements; conduct model evaluation with rigorous benchmarking methodologies (including symbolic verification layers). 
• Research-to-Production. Own the full lifecycle from exploratory research and paper publication through system design, A/B testing, and deployment on scalable cloud-native ML infrastructure. 

Qualifications

Education 

• Ph.D. in Formal methods/logical solvers, ML/DL, NLP, Computer Science, AI, Cognitive Science, or a closely related field — or a Master's degree plus 5+ years of commensurate industrial research experience. 

The perfect candidate will have the following technical skills

Domain: Depth Required 

Generative AI / LLMs Expert: transformer architectures, RLHF/DPO, prompt engineering, RAG, tool use.

Machine Learning Proficient: supervised/unsupervised learning, representation learning, Bayesian methods. 

Neuro-Symbolic Integration Proficient: at least one of LTN, NeSy-PI, DeepProbLog, NSCL, LARK, or own novel approach. 

Knowledge Graphs Proficient: Neo4j, Amazon Neptune, or Azure Cosmos DB for Gremlin; graph neural networks 

Research Record 

• At least 3 peer-reviewed publications in top-tier ML/SE/CS venues (e,g, PLDI, ASE, FSE, NeurIPS, ICML, ICLR, ACL, EMNLP, ICAPS, IJCAI, KR, AAAI, ESWC, ISWC, or equivalent). 
• Demonstrated ability to translate research outcomes into product or platform impact. 

Engineering Practices 

• Production Python, strong software design, cloud tools; 
• Experience with AI coding tools (Claude Code, Cursor)