From Data Platform Battles to AI Dreams: A Senior Engineer’s 3‑Year Journey at Alibaba

1. When a Project Is Critical, Just Do It

During the past three years at Alibaba, the author led a 5K+ generic big‑data platform project that spanned two regions and multiple teams. By stepping forward in a heated discussion, he helped accelerate responsibility allocation, which was resolved within 20 minutes, earning emotional reactions from senior leaders.

He later managed two small modules, quickly fixing issues and earning the “Best Firefighter” award, while observing Alibaba’s project‑management practices.

2. It’s Not Just My Work – Striving for Perfection

After the 5K+ platform, a critical module became unstable. The team decided to replace it with a new design, temporarily taking over some business work to free resources. The new module evolved into a data service platform that deepened interaction with the data team.

The author emphasizes that the success was a collective effort, creating lasting memories for all participants.

3. Technical Challenges as Opportunities

The author describes a typical challenge: implementing a rate‑limiting service that caps QPS per tenant. He outlines problem scoping, considering network attacks, future scale, stability, and architectural extensibility.

Choosing a solution involved distributed considerations (CAP theorem) and adopting a BASE‑style model with partitioned quota pools for large tenants and a separate mechanism for tail tenants, while also planning for failover, degradation, and scalability.

The overall lesson is to define clear goals, metrics, observability, evaluability, extensibility, recoverability, and hand‑offability.

4. No Need to Switch to Management at 35

The author reflects on mentorship, continuous learning, and his strengths in backend engineering and data platform construction. He is actively building knowledge in artificial intelligence, especially natural‑language processing, revisiting probability and statistics, and aiming for deeper theoretical breakthroughs rather than quick paper publications.

He argues that moving into management is not a necessity; effective management, like coding, requires theory and practice, and should not be forced upon engineers who are passionate about technical work.

He spends about three hours each day on machine‑learning study during his commute, while dedicating work hours to business code development, expressing the joy of solving technical problems.

He concludes with a personal note about staying true to the “technical heart” despite the ups and downs of corporate life.

Author: He Zhaowei, Senior Technical Expert at Alibaba.