AI Engineer - Platform

Crisil; Greater London; On-site; GBP 80,000 - 120,000; Full time

Job summary

CRISIL is seeking a seasoned ML/AI engineer to build and enhance AI-powered features within our products, focusing on production readiness and end-user value. You will collaborate with data scientists, AI engineers, and platform teams to deliver scalable AI capabilities.

Responsibilities include developing agent tools, optimizing prompts, and implementing testing, observability, and release criteria. Strong Python skills and experience with MLflow or similar tools are essential.

Qualifications

• 3+ years of experience delivering production ML/AI features.
• Strong Python proficiency for building and testing AI/ML applications.
• Experience with ML lifecycle: experimentation, evaluation, and release management.

Responsibilities

• Build and enhance AI-driven features within existing products.
• Develop agent tools, connectors, and integrations for AI workflows.
• Refine prompts and interaction patterns for production use cases.
• Support experiment tracking and model/version management.
• Contribute to quality gates, automated checks, and release readiness.

Skills

ML/AI engineering

Python

Experiment tracking

Model/version management

Communication skills

CI/CD workflows

Tools

MLflow

Git

Job description

Build, enhance, and maintainAI-powered product featureswith a focus oncontinuous product enhancement (CPE)and measurable end-user value. Engineer production-ready AI capabilities (agent tools, prompt optimisation, evaluation and observability) while collaborating closely with data science, AI engineering, and platform stakeholders. Contribute to quality-first delivery throughtesting frameworks, automated checks, and model/feature validationpractices. This is a platform engineering and not infrastructure.

Key Responsibilities

• Build and enhance AI-driven features within existing products, prioritising incremental improvements and adoption.
• Develop and maintainagent tools, connectors, and integrations to enable AI workflows.
• Refine and optimisepromptsand interaction patterns for production use cases, including safety and reliability considerations.
• Supportexperiment tracking and model/version management(e.g., MLflow or equivalent).

Testing, Evaluation & Quality

• Build and maintain test functions for AI tools, including exploringagent-based testingpatterns (AI validating AI) where suitable.
• Develop evaluation frameworks for AI/agent performance (accuracy, robustness, regressions, hallucination controls as applicable).
• Implementquality gates, automated checks, and release readiness criteria for AI features.
• Supportchampion/challengervalidation approaches prior to promotion.
• Contribute tomodel observabilityand dashboards to track quality, performance, and usage patterns.
• Partner with Data Scientists and AI Engineers to productionise research outcomes into reliable product capabilities.
• Coordinate with Platform/DevOps stakeholders for deployment needs and runtime requirements (without owning infrastructure delivery).
• Support tactical AI solutions in response to emerging business requirements.
• Participate in code reviews, documentation, and knowledge-sharing to strengthen team engineering standards.
• Work within agile/team delivery practices and contribute to shared sprint and release goals.

Skills required:

• 3+ yearsof experience in ML/AI engineering with demonstrable experience delivering production features.
• Strong proficiency inPython(designing, building, testing, and maintaining AI/ML applications).
• Understanding of theML/AI lifecycle, including experimentation, evaluation, and release management.
• Experience withexperiment trackingand model/version management concepts (tools such as MLflow are a plus).
• Strong communication skills with the ability to work effectively across technical and business stakeholders.

Preferred

• Hands-on experience withMLOps tooling(e.g., MLflow).
• Exposure toAI/agent frameworks(e.g., LangChain, LangSmith, CrewAI, Azure AI Agent Service, AWS Strands or similar).
• Experience withGenAI application development, including prompt iteration and agent/tool patterns.
• Familiarity with cloud platforms (AWS/Azure) and/or data platforms (e.g.,Databricks).
• Working knowledge ofGitandCI/CDworkflows.
• Experience infinancial servicesor other regulated environments.

Supporting Tools (Working Knowledge)

• Python ecosystem for AI/ML development (testing, packaging, experimentation)
• Git-based development workflows and code review practices

Observability/monitoring dashboards for AI feature performance (tooling varies by environment)