Engineer working at the intersection of applied AI, data-intensive systems, and pragmatic product development. I care about robust design, clear architecture, and using ML where it actually improves outcomes.
- 💼 Several years building and operating production-grade software systems, from user-facing experiences to data- and AI-powered backends
- 🎓 Master’s specialization in Machine Learning, with research on vision models and deep reinforcement learning
Applied Machine Learning
- Supervised and unsupervised learning for real-world tabular, text, and image data
- Vision models for segmentation, detection, and understanding of technical drawings and AEC data
- Evaluation design, experiment tracking, and closing the loop between model metrics and business impact
AI Systems & Agents
- Multi-step, tool-using “agentic” workflows that orchestrate multiple services instead of single prompts
- Retrieval and context construction pipelines that ground models in domain data
- Human-in-the-loop feedback, guardrails, and monitoring for AI features in production
Software & System Design
- Backend and API design for scalable, maintainable services
- Frontend integration that keeps AI features understandable and controllable for users
- Architecture-first thinking: decomposition, boundaries, and evolvable designs rather than one-off scripts
Data Engineering
- Batch and streaming-style data processing for analytics and ML
- Lakehouse-style data architecture, medallion (bronze/silver/gold) patterns, and versioned datasets
- Data modeling and interoperability between interactive notebooks, dataframes, and distributed compute
- 💼 LinkedIn: https://www.linkedin.com/in/uzair-sipra-85a961139
- ✉️ Email: sipra@ualberta.ca