About

• Created and architected the framework on which AI agents across Data, Engineering and business teams are built, deployed and observed. It grew from my own repository into a company-wide standard with contributions from more than 60 engineers.
• Built Fever’s first production LLM agent for external customers: an analytics assistant that lets venues and event organisers query sales and performance data in natural language inside the partner portal. Led it from prototype through production and partner rollout.
• Took the company-wide knowledge platform from proposal to production in under a month, combining retrieval, access control and human curation to ground agents in Fever’s business context.
• Helped set AI technical direction through architecture proposals, build-versus-buy decisions, agent harnesses, agentic development and incident response. I also design and run hiring loops, mentor a university project team, and train the Data organisation on agent development.

• Core member and day-to-day coordinator of a rapid-prototyping squad exploring AI products and taking promising ideas into production.
• Built Fever’s first agent-driven sales channel: a conversational planning and booking assistant on Telegram, with purchases completed inside the conversation.
• Implemented the pricing and payments layer using Stripe and the ticketing API, together with the checkout mini-app.
• The codebase I started during this work became the foundation of Fever’s AI agent platform.

• A leading contributor to dynamic pricing: weather-based pricing, a rule-simulation engine, partner promotions, and the multi-country pricing programme.
• Modernised the platform towards an event-driven architecture using Kafka and Redis, and served as a designated code reviewer.
• Built recommendation and gift-card pipelines, and fixed and generalised critical production processes.
• Shipped my first production feature three weeks after joining.

• Designed and productionised Python ML libraries and services, establishing packaging, testing, linting and CI/CD practices for reliable delivery.
• Built ingestion and transformation workflows using orchestration and lakehouse/warehouse patterns, improving data availability and reliability for ML use cases.
• Implemented ML APIs and serverless deployments, balancing performance, cost and operational stability.
• Delivered custom ETL workflows across cloud services and databases, improving scalability and reducing operational overhead.
• Led a database migration and cost optimisation initiative, and introduced tools and code-quality practices that raised engineering maturity.

• Delivered automation solutions using network orchestration platforms and tooling for large-scale service provider environments.
• Worked directly with enterprise customers to turn requirements into deployable technical designs and reliable workflows.
• Co-created and taught an internal Python workshop for engineers.
• Took part in applied AI and automation research and enablement initiatives.

• Led structured troubleshooting and incident-resolution workflows for enterprise networking systems in the Enterprise LAN Switching TAC team.
• Developed customer communication habits around expectation-setting, executive summaries and clear technical explanations.
• Completed technical training and certifications while supporting customers in production.

• Delivered data and ML initiatives from the initial business question to working systems within a new data function.
• Built recommendation and risk/fraud-related models to improve user experience and operational trust.
• Worked with Product and Operations to turn business questions into measurable data products.

Built features for iOS and Android using Ionic and Angular, collaborating with Product and Design from implementation through release.