Diogo Alves

• Own technology, development and infrastructure across the company — accountable for technical strategy, the engineering team, and everything running in production
• Own the product and application roadmap alongside the founders, turning business goals into technical milestones
• Lead and coach a cross-functional team of five engineers across backend, frontend, mobile and infrastructure — technical direction, code review, and growing the team's practice
• Architect and build the platform end to end: Python/FastAPI services, a TypeScript monorepo spanning web, mobile and internal apps, PostgreSQL, and infrastructure as code on Azure
• Set and enforce the engineering quality bar: pull-request review on every change, unit, integration and end-to-end test suites, and CI gates that block merges on failing tests, linting or architectural violations
• Design for privacy by default when handling sensitive personal data — data minimisation, pseudonymisation and hashing of identifiers, least-privilege access, managed secrets and network isolation, aligned with GDPR
• Design and ship production LLM agent systems — tool use, guardrails, streaming and long-term memory — with architectural boundaries enforced in CI
• Introduced AI engineering rigor: automated evaluation pipelines, model-selection benchmarking, latency budgets and explicit rollout gates
• Own the engineering hiring process end to end, from technical assessment design to onboarding

• Led AI applications for finance — NLP systems, RAG chatbots and AI agents — from prototype through to production
• Owned the MLOps platform on Azure Machine Learning: training and deployment pipelines, model registry, monitoring and CI/CD
• Deployed and operated production AI services on AKS, responsible for reliability and performance in a regulated environment
• Set engineering standards for the AI workstream — pull-request code review, automated testing and reproducible pipelines — and mentored engineers on ML and MLOps practice
• Partnered with business stakeholders to scope use cases and translate them into ML systems that shipped
• Worked within the data governance and privacy constraints of a regulated financial environment, handling sensitive data under strict access and audit requirements

• Taught data science and machine learning to working professionals in an intensive career-change programme
• Delivered the Analyzing and Visualizing Data and Machine Learning Models modules, from fundamentals to hands-on practice
• Part of a consistent thread of teaching and coaching that carries into how I lead engineering teams

• Built machine learning for industrial use cases: time series forecasting and anomaly detection, and computer vision for monitoring and automated inspection
• Owned the full model lifecycle — data collection and labelling, feature engineering, training, evaluation and deployment
• Took models from research prototype to deployed service, containerised and served through APIs into client systems
• Where the ML engineering thread started — a continuous line running through every role since

• Built and operated production data pipelines with Spark, Airflow, Hadoop and Hive for large-scale batch processing
• Designed data models and orchestration for downstream analytics consumers
• Where the engineering fundamentals came from — distributed systems, data modelling and pipelines at scale

• Researched machine learning for anomaly and burst detection in flow-measurement time series
• Published as an open-source project (WISDom) — first end-to-end ML system, from problem framing to evaluation

Education

Continuous learning journey