1 to 25 of 51 Permanent Human-in-the-Loop Jobs in the UK

Prompt and few‐shot experimentation Hallucination detection and mitigation Deterministic and rule‐based validation SME‐driven constraints and edge‐case testing Human‐in‐the‐loop feedback mechanisms Rapidly prototype under uncertainty, independently converting loosely defined business questions into functioning AI tools with minimal supervision and tight ...

rollout/rollback), guardrails, telemetry, drift monitoring, prompt safety. DevSec & Responsible AI: Privacy‐by‐design, PII handling, secrets management, policy enforcement, human‐in‐the‐loop approvals, auditability. Languages: Advanced Python plus one of Java/TypeScript/Go; containerization (Docker/K8s). Leadership & Communication Executive ...

deliver pragmatic AI/ML opportunities where valuable (e.g., agentic workflows, RAG/LLM-enabled capabilities) with appropriate governance and human-in-the-loop controls. What you need to have: Experience managing a team, as this role involves leading a team of 5 as well ...

Experience deploying to AWS (e.g., SageMaker, Bedrock) and operating production workloads with attention to cost, performance, security, and scaling. Experience integrating human-in-the-loop or user feedback signals into iterative improvement processes. If you’re ready to make a lasting impact in a fast ...

workflow orchestration - across multiple business verticals Architect, prototype, and productionize agentic pipelines: agent orchestration, memory and context management, tool integration, and human-in-the-loop governance frameworks Define agent evaluation frameworks covering reliability, hallucination risk, task completion rate, and regulatory-grade auditability Design and implement the ...

agent engineering: Build and tune LLM workflows, prompt strategies, schema-driven extraction, tool-calling patterns, agent orchestration, evaluation loops, and human-in-the-loop controls. Enterprise integration: Integrate AI solutions with enterprise systems, APIs, data platforms, document repositories, workflow tools, observability platforms, and identity and access ...

complex trade offs, and synthesis of large amounts of technical and regulatory information to inform organisational decision making.3. Define and support human-in-the-loop processes for high-risk AI systems to ensure accountability and intervention capability including prolonged assessment of model risk profiles, failure modes ...

enterprise-authorized AI capabilities within the work environment to accelerate architecture analysis and decisioning across the product family, with human-in-the-loop validation and appropriate handling of sensitive data. Establishes reuse-first AI-enabled engineering patterns and governance across SDLC/toolchain practices ...

compute, CoT pipelines, Tool use & API calling, etc.) Data-centric Machine Learning (Synthetic Data, Curriculum Learning, Learned datamixtures, etc.) Evaluation (Benchmarks, Human-in-the-loop, red teaming/Adversarial Testing, Hallucination detection, ...) We work collaboratively both with TR Labs (TR’s applied research division), academic ...

identified and controlled throughout the product lifecycle (including logging/traceability, data residency, third party risk, access/entitlements, and human in the loop safeguards where applicable), challenge and influence teams to remediate gaps, and escalate through governance channels or other mechanisms when needed ...

identified and controlled throughout the product lifecycle (including logging/traceability, data residency, third party risk, access/entitlements, and human in the loop safeguards where applicable), challenge and influence teams to remediate gaps, and escalate through governance channels or other mechanisms when needed ...

trust an automated system with high‐stakes decisions, which means designing for explainability, evidence and provenance, calibrated confidence, and clear human‐in‐the‐loop control. You'll shape how AI‐generated findings are surfaced, sourced and acted on — turning model complexity into clarity people can rely ...

Foundry ecosystem, understanding different business challenges require varied solutions from SLM s to intelligent Agentic automation from LLM foundational models, Human in the Loop and MLOps. Data Architecture Design: You understand and know when to use the following data architecture design patterns RAG (Retrieval-Augmented ...

security controls, access management and solution assurance Principles of prompt engineering and agent orchestration; appreciation of grounding data, hallucination mitigation, and human in the loop controls Demonstrable high level of understanding and working practice of the processes, regulations and codes of practice (ethics/ ...

operation of agentic AI capabilities Define how agents interact with users, CRM data and workflows, including boundaries, controls, escalation paths and human-in-the-loop mechanisms Ensure agentic behaviour is understandable, predictable and trustworthy through strong guardrails, safety mechanisms and observability Contribute and partner on core ...

enabled development patterns (using enterprise-authorized tools within the work environment) across teams/functions, including establishing governance for human-in-the-loop validation, traceability/auditability, and secure handling of sensitive inputs/outputs. Strong understanding of responsible AI use and control expectations ...

Artificial Intelligence (AI) pipelines, including feature engineering, model training, evaluation and monitoring Experience designing evaluation frameworks for LLM use cases (e.g., human-in-the-loop, calibration, guardrails, auditability) Experience with hyperparameter tuning at scale and experimentation frameworks Strong technical experience with SQL and Python; significant experience ...

compute, CoT pipelines, Tool use & API calling, etc.) Data-centric Machine Learning (Synthetic Data, Curriculum Learning, Learned datamixtures, etc.) Evaluation (Benchmarks, Human-in-the-loop, red teaming/Adversarial Testing, Hallucination detection, ...) We work collaboratively both with TR Labs (TR’s applied research division), academic ...

identified and controlled throughout the product lifecycle (including logging/traceability, data residency, third‐party risk, access/entitlements, and human‐in‐the‐loop safeguards where applicable), challenge and influence teams to remediate gaps, and escalated through governance channels or other mechanisms when needed ...

enabled development patterns (using enterprise‐authorized tools within the work environment) across teams/functions, including establishing governance for human‐in‐the‐loop validation, traceability/auditability, and secure handling of sensitive inputs/outputs. Strong understanding of responsible AI use and control expectations ...

working with business or domain data to ground, evaluate and improve AI outputs. Ability to design feedback loops, evaluation approaches and human-in-the-loop processes to improve AI solution quality over time. Strong understanding of responsible AI considerations, including privacy, security, hallucination risk, bias, transparency ...

Experience leading multi-team adoption of enterprise-authorized AI-assisted development and delivery tools, including defining governance/ways of working (human-in-the-loop validation, quality gates), measuring outcomes, and ensuring secure handling of sensitive inputs/outputs. Strong understanding of responsible AI use in ...

Experience with AI tools such as Claude, ChatGPT, Microsoft Copilot or similar LLM platforms. Familiarity with AI-enabled workflow automation and human-in-the-loop agentic solutions. Knowledge of enterprise data platforms, automation tools, and service delivery technologies. Experience collaborating with product, engineering, data and platform ...

migration from legacy infrastructure to contemporary architectures. Awareness of AI and automation opportunities within clinical and operational settings, including clinical risk, human-in-the-loop design and regulatory considerations. 4. PERSONAL ATTRIBUTES Deeply curious about healthcare operations and motivated by improving clinical and business outcomes rather ...

assisted analysis patterns. Familiarity with sophisticated AI workflow techniques such as agentic workflows, task decomposition, evaluator/critic loops, human‐in‐the‐loop review, context or memory systems, and traceable evidence generation. Ability to work with domain experts, extract tacit knowledge, and translate it into structured ...