At ciandt, we are experts in technology transformation, combining human talent with artificial intelligence to develop scalable technological solutions.With a global team of over 8,000 professionals, we have established partnerships with over 1,000 clients during our three decades of operation. Artificial Intelligence is at the core of what we do.MissionAs a Principal Architect, your primary goal will be to spearhead the execution and delivery of AI engineering projects. You will ensure adherence to engineering standards, alignment with business goals, and the delivery of tangible value through the integration of advanced AI systems.Key ResponsibilitiesAssist the Engineering Manager with daily operations.Oversee the daily execution of project delivery, including planning, managing dependencies, risk assessment, and ensuring commitments are fulfilled.Guarantee the application of engineering standards, including defining completion criteria, testing strategies, code quality, documentation, and operational readiness.Communicate engineering updates to stakeholders clearly and escalate issues as necessary.Facilitate team execution without direct line management: guide onboarding processes and foster effective working practices, mentor engineers and tech leads, and share best practices.Ensure knowledge transfer and project handover: document decisions, runbooks, and key architectural choices to enable a smooth transition post-engagement.Collaborate with Product/Business teams to identify opportunities where agentic AI creates significant value (e.g., workflow automation, assisted decision-making, content acceleration).Translate business requirements into technical objectives (including latency, cost, quality, robustness, and compliance) and define success metrics (KPIs, A/B testing, guardrails).Design agentic architectures encompassing tool orchestration, planning, memory management, context handling, retrieval (RAG), routing, multi-agent patterns, and human-in-the-loop processes.Implement production patterns such as prompt/versioning, evaluation harnesses, regression tests, feature flags, and monitoring for quality, cost, and latency.Engineer systems for resilience: establish fallbacks, timeouts, retries, circuit breakers, safe tool execution, sandboxing, and secrets management.Implement guardrails including tool policies, content filtering, PII redaction, policy-as-code, access controls, auditability, and traceability (traces, conversations, decisions).Partner with IT/Security/Cloud teams to ensure compliance with privacy, security, and risk management at scale.Define and enforce quality gates prior to production, including red teaming, adversarial testing, and management of bias and hallucination risks.Develop and manage the machine learning value chain: data contracts, data quality, lineage, drift monitoring, dataset management, and training/inference pipelines.Oversee the production operations for models and agentic services, focusing on deployment, scaling, observability, incident response, SLOs, and post-mortem analysis.Industrialize continuous evaluation processes, including offline evaluations, golden sets, and feedback loops.
Apr 30, 2026