Job Specifications
We are seeking a visionary Lead AI Architect to define the technical strategy and end-to-end architecture for enterprise-scale Generative AI solutions. You will lead the design of autonomous Agentic AI systems and advanced RAG (Retrieval-Augmented Generation) architectures using the AWS ecosystem. This role requires a deep understanding of LLM lifecycle management—from prompt engineering and evaluation to fine-tuning and deployment—ensuring solutions are secure, scalable, and production-ready.
Key Responsibilities
· Solution Architecture: Design scalable, secure, and cost-effective architectures on AWS for Generative AI applications, leveraging Amazon Bedrock for model access and Amazon SageMaker for custom model hosting and fine-tuning.
· Agentic System Design: Architect complex, multi-agent workflows using LangGraph and LangChain that can reason, plan, and execute tasks autonomously while maintaining strict state management and guardrails.
· LLM Ops & Evaluation: Establish robust evaluation frameworks (e.g., RAGAS, TruLens) to measure model performance (faithfulness, answer relevance) and design "Human-in-the-Loop" workflows for continuous improvement.
· Fine-Tuning & Customisation: Lead strategies for domain adaptation, including PeFT (Parameter-Efficient Fine-Tuning) and Instruction Tuning on Amazon SageMaker to tailor open-weights models (e.g., Llama 3, Mistral) to specific business needs.
· Governance & Security: define standards for responsible AI, including data privacy, PII redaction, and protection against prompt injection, utilising AWS Guardrails, Amazon Comprehend and IAM policies.
Technical Skills & Requirements
· Core Cloud: Expert-level knowledge of AWS AI services (Bedrock, SageMaker, OpenSearch Vector Engine, Kendra).
· Generative Frameworks: Mastery of LangChain, LangGraph, and Python-based orchestration layers for building stateful agents.
· Model Engineering: Deep understanding of Transformer architectures, embedding models (Titan, Cohere), and techniques for Fine-Tuning and RLHF (Reinforcement Learning from Human Feedback).
· Development Standards: Strong background in software architecture patterns (Microservices, Event-Driven Architecture) and IaC (Terraform/CloudFormation).
Advantageous Skills
· Middleware Integration: Experience with Workato or similar iPaaS platforms to orchestrate workflows between AI agents and enterprise applications (e.g., SAP, Salesforce) is a distinct advantage.
· Data & Traditional ML: Proficiency with Databricks (Lakehouse, Unity Catalog) and traditional ML algorithms (Regression, Forecasting) is highly desirable, enabling hybrid solutions that combine predictive analytics with generative capabilities.
· Any experience in Energy & Utilities industry is advantageous.
About the Company
Deltaclass Technology Solutions (DCTS) is an IT Recruitment and Consultancy Services Company, and we are specialised within Technology hiring.
- Supply Permanent / Temporary / Project based staffing Solutions.
- Our Recruitment Consultants are very well trained to fulfil our clients need.
- Operational global presence in UK, Germany, Netherlands, Sweden, Poland with operations across Europe, India and USA.
Know more