AI Architect – Agentic AI & LLM Platforms

Company Description :
BAPON Technologies is a forward-thinking IT consulting company that helps businesses achieve intelligent automation, seamless digital integration, and effective cloud transformation. As a Cyber Essential Plus and ISO 9001 & ISO 27001 certified organization, BAPON maintains high standards of quality, security, and compliance for clients, particularly startups and emerging companies. The team combines deep expertise in low-code, RPA, AI, and ML to streamline operations and improve productivity, while also enabling robust integration of systems, applications, and data. BAPON supports clients through strategic IT planning and tailored cloud solutions that drive innovation and growth. The company is also building SkillfulSense, an AI-powered skill assessment platform designed to help learners identify strengths, close skill gaps, and grow with confidence.
Role Description:
We are looking for an experienced AI Architect to lead the architecture and engineering of enterprise-grade Agentic AI and LLM solutions.
The role will be responsible for defining the architecture, engineering standards, governance, and quality framework for AI agents across the platform. You will provide technical leadership across agent orchestration, prompt engineering, RAG, tool integration, evaluation, security, performance, and production deployment.
This is a senior hands-on architecture role requiring strong experience designing and delivering production AI/ML and Generative AI solutions.
Key Responsibilities:
• Define the overall Agentic AI architecture, including agent orchestration, multi-agent patterns, memory, context management, tool calling, and workflow integration.
• Establish engineering standards and reusable patterns for building, testing, deploying, and operating AI agents.
• Lead architecture and design reviews for AI agents and Generative AI solutions developed across the engineering teams.
• Architect Retrieval-Augmented Generation (RAG) solutions, covering document ingestion, chunking, embeddings, vector search, retrieval strategies, reranking, grounding, and response generation.
• Define standards for prompt engineering and prompt lifecycle management, including versioning, testing, governance, and reusable prompt patterns.
• Design the AI evaluation framework, including golden datasets, hallucination testing, groundedness, relevance, accuracy, confidence scoring, regression testing, and human evaluation.
• Define integration patterns between AI agents, enterprise APIs, databases, applications, and external tools.
• Establish standards for Model Context Protocol (MCP) and other tool/function-calling approaches where appropriate.
• Provide architectural guidance on model selection and integration across commercial and open-source LLMs.
• Define guardrails for responsible AI, security, data privacy, access control, observability, and auditability.
• Monitor and optimise production AI workloads across accuracy, latency, reliability, token consumption, and cost.
• Provide technical leadership and mentoring to AI/ML engineers and support the development of engineering capability across the team.
• Work closely with platform, cloud, data, security, architecture, and product teams to ensure AI solutions integrate effectively into the wider enterprise technology landscape.
Required Skills:
• 10+ years of experience across AI/ML, software engineering, data engineering, or solution architecture, with significant recent experience delivering production Generative AI/LLM solutions.
• Strong hands-on knowledge of Agentic AI architectures, including agent orchestration, reasoning workflows, tool/function calling, memory and context management.
• Deep understanding of LLMs, prompt engineering, model integration, and Generative AI application architecture.
• Proven experience designing and implementing enterprise RAG architectures.
• Strong understanding of embeddings, vector databases, semantic/hybrid search, reranking, retrieval optimisation, and grounding techniques.
• Experience designing LLM and agent evaluation frameworks, including hallucination detection, quality metrics, golden datasets, confidence assessment, and regression testing.
• Strong Python engineering capability.
• Experience with AI/LLM frameworks and SDKs such as LangChain, LangGraph, Semantic Kernel, LlamaIndex or equivalent technologies.
• Strong understanding of REST APIs, event-driven architectures, microservices, and enterprise integration patterns.
• Experience deploying AI solutions into production cloud environments such as AWS, Azure or GCP.
• Understanding of AI observability, monitoring, security, governance, and responsible AI practices.
• Experience managing LLM cost, token consumption, latency and performance optimisation.
• Experience providing technical leadership across AI/ML engineering teams.
Nice to have:
• Experience with Model Context Protocol (MCP) and emerging agent interoperability standards.
• Experience designing multi-agent systems and agent-to-agent communication patterns.
• TypeScript/Node.js development experience.
• Experience with vector databases and search platforms such as Pinecone, Weaviate, OpenSearch, Elasticsearch, Azure AI Search, pgvector or equivalent.
• Experience with enterprise AI platforms and LLMOps/MLOps.
• Knowledge of ITSM, IT operations, service management or enterprise automation domains.
• Experience integrating AI agents with platforms such as ServiceNow, Salesforce, Microsoft, or other enterprise applications.
• Understanding of AI security topics including prompt injection, data leakage, model access controls, content filtering, and agent/tool permission management.
Key Outcomes:
The AI Architect will ensure that the organisation has a consistent, scalable and governed architecture for Agentic AI, enabling engineering teams to build AI agents that are accurate, secure, observable, cost-effective and suitable for enterprise production use.

Work arrangement
No