Senior AI Engineer

About the Role

As a Senior AI Engineer, you will take end-to-end ownership of services on our AI Platform — a serverless, AWS-native platform that powers AI and machine learning across insurance operations. The platform supports both generative AI services (LLM-based document processing, extraction, classification, RAG, and chatbots) and custom ML models trained in-house (SageMaker pipelines, classical ML, and computer vision), served through a shared, governed infrastructure. You will make sure the key technical and design decisions — spanning Terraform infrastructure, CI/CD, model training and integration, inference, monitoring, and downstream integration — while ensuring that it meets our AI governance and Responsible AI standards.

You will lead the delivery of new AI initiatives across the organization — including fraud detection, document processing and intelligent extraction, claims and underwriting automation, and AI for operations — driving each from requirements and architecture through to a production, monitored deployment. As one of the experienced engineers on the team, you will set the technical direction and quality bar through hands-on design and code review, and help other engineers grow.

Responsibilities

1. Solution architecture ownership. Own the solution architecture for AI services and new initiatives — make and document technology and design decisions, define integration patterns across services, and ensure solutions align with platform standards, security, and AI governance.
2. Technical leadership & mentoring. Mentor and support other engineers, share knowledge across the team, and help grow AI/ML capability. Set and uphold engineering best practices through code and architecture review.
3. Architecture & code review. Review designs and code from the team to ensure quality, maintainability, security, and consistency with platform standards, and provide constructive, actionable feedback.
4. Deliver new AI projects. Take new initiatives from concept to production across areas such as fraud detection, document processing and intelligent extraction, underwriting automation, and AI for operations. Gather requirements, design the solution and architecture, build it on the shared platform, and hand it over as a monitored, governed production service.
5. Platform engineering & IaC. Implement and enhance the AI and analytics platform using Infrastructure as Code with Terraform. Keep changes aligned with existing VPC, IAM, API Gateway, and data-layer architecture.
6. End-to-end ML/AI lifecycle. Manage models across the full lifecycle. This covers training and deploying custom ML models (e.g. classification, regression, object detection / computer vision) via our SageMaker MLOps pipeline (preprocess → train → evaluate → conditionally register), as well as integrating LLM as needed. Cover model import, training, evaluation, deployment, inference execution, monitoring, and integration of outputs into downstream applications and automation workflows.
7. Serverless AI pipelines. Build scalable, serverless AI and data processing pipelines on AWS.
8. Generative AI solutions. Develop and integrate GenAI features — LLM-based classification and extraction, document/case analysis, RAG, and chatbots — into operational workflows.
9. Automation of core insurance operations. Support automation across: (i) data capture and verification from documents (OCR + LLM-based extraction); (ii) automated adjudication, eligibility checks, and benefit/reserve calculations; (iii) fraud/anomaly detection using ML and analytics; and (iv) workflow automation for processes such as claims assessment, payment initiation, notifications, and underwriting.
10. Vendor evaluation & third-party integration. Support end-to-end vendor evaluation and procurement (RFI, RFP, selection, due diligence) and adapt third-party AI solutions into the GIMB environment, aligned with architectural principles, AI governance, and technical standards. Manage integrations with enterprise platforms, data services, and corporate systems.
11. Cross-functional collaboration. Work with data engineers, business analysts, business units, and other stakeholders to gather requirements, onboard new use cases and lines of business, and integrate AI outputs into business processes and reporting.
12. AI governance & Responsible AI. Support the AI governance framework: complete AI Risk Assessments before production deployment, maintain the AI Model Registry, document prompt designs and model/classification logic, keep human-in-the-loop QC workflows, and enforce data-privacy controls.
13. Ground truth, evaluation & observability. Develop ground-truth datasets and labeling processes for training, evaluation, and monitoring. Compile accuracy/performance metrics and statistical analysis to drive model improvement. Implement structured logging, CloudWatch dashboards, and alarms/SNS notifications for service health, error rates, and model degradation.

Requirements

Must-have

· Proficiency in Python and ML frameworks/algorithms (e.g. scikit-learn, XGBoost, PyTorch, TensorFlow).

· Strong grasp of generative AI and LLM techniques: prompt engineering, structured extraction, RAG, embeddings, and integrating hosted models (Anthropic Claude, Amazon Bedrock).

· Experience training and operationalizing custom ML models end to end (data prep, training, evaluation, deployment, monitoring).

· Experience designing and deploying on AWS, hands-on with core services used here: Lambda, API Gateway, Step Functions, SageMaker, Bedrock, Textract, DynamoDB, S3, ECR, SSM Parameter Store, CloudWatch, VPC/IAM.

· Infrastructure as Code with Terraform — modules, per-environment state backends, plan/apply workflows (AWS CDK a plus).

· Solid MLOps understanding: CI/CD, model registration, monitoring, and drift detection.

· Proven ability to own solution architecture and mentor engineers, including leading design and code reviews.

· Familiarity with event-driven and serverless architectures (Lambda / Step Functions).

· Container skills: Docker (Python slim base images, Lambda container images) and image lifecycle management in ECR.

· CI/CD with GitHub Actions, including OIDC federation to AWS (no long-lived credentials) and environment-gated promotion.

Nice-to-have

· OCR and document-intelligence experience (AWS Textract, IDP pipelines).

· Computer vision / object detection experience (e.g. training and deploying detection models).

· Vector search / RAG with a vector store (e.g. pgvector).

· Data warehouse familiarity such as Amazon Redshift.

· Testing and quality tooling: pytest, property-based testing (Hypothesis / fast-check), and SAST/dependency scanning.

· API contract discipline with OpenAPI specs.

· Full-stack awareness for AI applications: FastAPI backends and React/Vite/MUI frontends.

· Exposure to Responsible AI practices and LLM guardrails (content filtering, confidence scoring, human review).

· AWS certifications (AI/ML Specialty, Solutions Architect) or equivalent.

Interested candidates, please share your updated CV or get in touch for more details.

Email: [Confidential Information]

WhatsApp: 60122456834