Denis Sineiko

• Designed and built a production Python backend and automation platform from proof of concept, establishing reusable architecture across 30+ external lender platforms.
• Support thousands of weekly workflow executions, submissions and data extraction operations, processing multiple cases concurrently to reduce adviser wait time and manual workload.
• Developed workflows for data retrieval, validation, submissions, document processing and portal uploads, replacing manual tasks taking approximately 15 minutes to 2 hours per case.
• Designed, built and maintained production REST APIs using Flask and FastAPI for workflow automation, external integrations, callbacks, file operations, validation and error handling.
• Built production observability with Sentry, structured logging, health checks and workflow-level diagnostics to investigate API failures, timeouts and application/container incidents.
• Designed an API-driven regression and validation framework covering approximately 50 to 60 end-to-end scenarios, using API checks, pytest, Selenium validation, screenshot evidence and failure diagnostics.
• Own technical delivery across backend architecture, automation development, releases, dependency maintenance and production support, including performance and reliability improvements.
• Integrated AI/LLM services for document and data extraction, applying structured output validation, inconsistency analysis, cross-model verification and deterministic fallback logic.
• Established coding standards and reusable integration patterns; provide code reviews, technical guidance and solution design support within a small product team.
• Maintain project-specific Docker images and GitLab CI/CD workflows with Pylint checks; collaborate with DevOps on AWS/ECS deployments and environment-level issues.

• Automated reconciliation between banking and retail datasets, achieving approximately 96% automatic matching and reducing manual work to exception handling.
• Built a reusable Python ETL and data collection framework from scratch across 16+ sources, including APIs, databases, files and business data feeds.
• Cleaned and transformed business and global sales data for BigQuery, developing transformation logic, views and reporting datasets used across the organisation.
• Created and supported 20+ Looker and Data Studio dashboards, translating operational data into metrics for teams and individual stakeholders.

• Automated daily data cleaning, transformation, template mapping and Excel report generation, reducing routine processing time by approximately 50%.
• Optimised queries and processing logic across large datasets while maintaining data accuracy from raw input through final production output.

• Built Python prototypes and project workflows using asyncio, async/await and HTTPX for concurrent API and integration workloads.
• Developed practical LLM-enabled workflows using OpenAI, Anthropic and Gemini APIs, including document extraction, speech-to-text processing and AI-assisted automation.
• Built n8n workflows and lightweight AI agents with tool calling for chatbot requests, information extraction and response suggestions, using structured outputs and validation around model responses.

Occasionally volunteer my time to help people in my community with technology and practical everyday tasks, particularly where professional help or replacement equipment may be difficult to afford.