Role Description We are looking for an Application Architect with strong AI engineering experience to design and build intelligent, agentic applications on Google Cloud Platform. This role sits within an application engineering team and focuses on architecting AI-enabled systems using the Google Agentic Development Kit (ADK), Gemini, and Vertex AI — integrated into enterprise Java/Python backends and cloud-native microservices. You are equally comfortable defining application architecture, designing agentic workflows, writing production-quality code, and translating AI capabilities into practical, mission-aligned solutions for federal stakeholders. This role is remote with a preference for candidates located in Virginia, Maryland, or Washington DC. Key Responsibilities - Architect and implement AI-enabled application systems on GCP, with a focus on agentic workflows using Google ADK and Gemini Pro. - Design human-in-the-loop agentic systems — defining agent roles, tool use, orchestration patterns, and guardrails for responsible, auditable AI behavior. - Integrate AI/ML capabilities (Vertex AI, Gemini APIs, embeddings, RAG) into enterprise Java and Python applications via well-designed APIs and microservices. - Lead application-layer design decisions: data flow, context management, session handling, and state management within agentic architectures. - Collaborate with Data Engineers (BigQuery, Dataform) and Cloud Architects to ensure AI application solutions are grounded in reliable, governed data. - Conduct architectural reviews, define coding standards for AI-integrated applications, and mentor engineers on agentic design patterns. - Evaluate AI use cases for feasibility, risk, and mission fit; prototype and validate approaches before committing to full builds. - Contribute to responsible AI practices: explainability, human oversight, auditability, and alignment with federal AI governance requirements. - Stay current on the Google AI ecosystem (Gemini, ADK, Vertex AI Agent Builder) and inform team and leadership on strategic direction. Qualifications - 5–8 years of software or application engineering experience, with demonstrated focus on AI-integrated or intelligent application design. - Hands-on experience with Google ADK or comparable agentic frameworks (LangGraph, LangChain, AutoGen); Google ADK strongly preferred. - Proficiency in Python for AI/ML integration; Java experience a plus in application team context. - Experience integrating LLM APIs (Gemini, OpenAI, or equivalent) into production application workflows. - Solid understanding of agentic design patterns: tool use, multi-agent orchestration, retrieval-augmented generation (RAG), memory and context management. - Experience with GCP services: Vertex AI, Cloud Run, GKE, BigQuery, Pub/Sub. - Familiarity with REST API design, microservices architecture, and CI/CD pipelines (Harness preferred). - Understanding of responsible AI principles: human-in-the-loop design, auditability, bias awareness, and federal AI governance. Desired Qualifications - Experience with Vertex AI Agent Builder, Gemini Code Assist, or Gemini CLI in a development workflow context. - Familiarity with GCP-native data tooling: BigQuery, Dataform, Looker. - Experience on federal or large-scale enterprise modernization programs. - Exposure to FedRAMP/FISMA requirements and security-compliant AI deployment practices. - Experience with DevSecOps pipelines (Checkmarx, Invicti, or equivalent SAST/DAST tooling). Additional Information - Successful completion of a client-required background investigation and suitability determination will be required. - The ability to obtain and maintain a federal security clearance may be required based on engagement. - Bachelor's degree in Computer Science, Software Engineering, or a related field; advanced degree a plus. - Google Cloud Professional Cloud Architect or Professional Machine Learning Engineer certification preferred. - Security+ desirable. Benefits - Health Care Plan (Medical, Dental & Vision) - Retirement Plan (401k, IRA) - Life Insurance (Basic, Voluntary & AD&D) - Paid Time Off (Vacation, Sick & Public Holidays) - Family Leave (Maternity, Paternity) - Short Term & Long Term Disability - Training & Development - Work From Home - Wellness Resources - Employee Bonus Programs

• Joining Customer Service Transformation Programme, working within the AI & Agents squad. • The squad owns customer-facing agentic AI on AWS Bedrock with multi-agent collaboration patterns using Claude Sonnet. • Concurrent work includes domain-specific chat agents going live, additional domain agents being designed, chatbot automation, and Agent WorkSpace AI features (real-time alerts, complaint indicators, Agent Note Taker).

Role Description We are looking for an experienced Senior Fullstack / AI Engineer who will be responsible for end-to-end feature development — from building intelligent AI agents on the backend to visualizing them through intuitive user interfaces. Why you’ll love this project: - We have absolutely zero legacy code. - Our entire stack is bleeding-edge (running the latest versions). - We don’t just "plug in an OpenAI API key" — we are building a complex distributed system featuring RAG architecture, vector search, event streaming, and big data analytics. Our Tech Stack: - Frontend: Vue 3 / React, TypeScript - Backend: Node.js (LTS) + NestJS (core business logic), Python + FastAPI (AI services, orchestration, data processing) - AI & ML: OpenAI, Anthropic, Google - Data Storage & Messaging: PostgreSQL, ClickHouse, Qdrant, Kafka, Redis, S3 / MinIO Key Responsibilities - Develop end-to-end functionality: from complex interactive interfaces and dashboards to distributed backend services using NestJS and Python. - Design and maintain a microservices architecture, ensuring seamless interoperability between Node.js and Python services. - Design and optimize data storage schemas across a heterogeneous environment (PostgreSQL, ClickHouse, Qdrant). - Implement asynchronous inter-service communication and event streaming via Kafka. - Integrate AI components, managing embeddings and semantic search functionality within Qdrant. - Optimize application performance, build times, and data processing pipelines. - Design and evolve the AI architecture: integrating OpenAI/Anthropic models, managing context windows, and optimizing token consumption (tiktoken). - Build and tune RAG (Retrieval-Augmented Generation) pipelines using LangChain and Qdrant vector database. Qualifications - Senior, 4–5+ years in product development, with hands-on experience in LLM integration and data engineering. - Strong Fullstack Background: Deep knowledge of Node.js, Worker threads, and the NestJS ecosystem (solid understanding of Dependency Injection, Modules, Guards, Interceptors, and Pipes). Proficient in TypeScript (Vue 3, NestJS) and a proven track record of backend development with Python. - Hands-on AI/LLM Experience: A solid understanding of prompt engineering, LangChain chains, embeddings, and semantic search. You know exactly how Claude differs from GPT in production environments. - Vector DB Expertise: Practical experience with Qdrant (or similar vector databases) — including indexing, similarity search, and vector tuning. - Data Infrastructure Literacy: Experience working with distributed systems, message brokers (Kafka), and analytical databases (ClickHouse). - Database Mastery: You clearly understand when to query PostgreSQL, when to aggregate data in ClickHouse (OLAP vs. OLTP), and how vector indexes operate in Qdrant. - Infrastructure Horizons: Experience with caching (Redis), object storage (S3/MinIO), containerization (Docker/Kubernetes), and the AWS ecosystem. - Mindset: Strong analytical skills, a product-driven mindset, the ability to architect fault-tolerant systems, and a dedication to writing clean, maintainable code (SOLID, OOP/FP patterns). Benefits - Consistent monthly payment based on invoices (payments are made via bank transfer/ Payoneer or PayPal). - Remote working from 14:00 to 23:00 EEST, including a one-hour break. - Supportive work/life balance: paid vacation and sick leave. - Opportunities for professional and career growth and development. - Opportunity to use modern approaches and tools to solve business problems. Our recruitment process - Interview with HR - Technical interview with team leaders

Role Description Joining Customer Service Transformation Programme, working within the AI & Agents squad. The squad owns customer-facing agentic AI on AWS Bedrock with multi-agent collaboration patterns using Claude Sonnet. Concurrent work includes: - Domain-specific chat agents going live - Additional domain agents being designed - Chatbot automation - Agent WorkSpace AI features (real-time alerts, complaint indicators, Agent Note Taker) Qualifications - 5+ years software engineering, with 2+ years on customer-facing AI/LLM systems in production - Hands-on Amazon Bedrock - Agent configuration, Action Groups, Knowledge Base integration, Guardrails - Multi-agent orchestration patterns (supervisor/collaborator/handoff) - Prompt engineering at production scale - versioning, A/B testing, regression methodology - Lambda fulfillment patterns including OpenAPI tool specifications - TypeScript primary, Python secondary - Working knowledge of vector databases (OpenSearch Serverless, Pinecone, or equivalent) - Strong written and verbal communication - this role contributes to architectural decisions visible to senior stakeholders Requirements - Hands-on Amazon Bedrock AgentCore (the new managed agent runtime - rare in market) - Amazon Agent Assist / Agent WorkSpace integrations - Voice agent experience (Nova Sonic or other voice LLM) - Production experience with Anthropic Claude on Bedrock (Sonnet 4 or later) - RAG implementation experience - chunking strategy, embedding model selection, retrieval evaluation - AI safety / Guardrails configuration experience Company Description AND Digital are a tech company focused on accelerating digital delivery and dedicated to closing the digital skills gap. We’ve been helping organisations build better digital products and stronger digital teams since 2014. - We believe our work should always make a remarkable impact for our clients. - This unique model has driven success for our clients and ourselves, evidenced by our remarkable organic growth since 2014. - Today we number more than 1,300 people with Clubs all over the UK, Europe and the USA with plans for global expansion in the next couple of years. Join us - and help us fulfil our mission to close the world’s digital skills gap.