AIGC SaaS and Model-serving Platform

Overview

At SenseTime AIGC, contributed to 0-to-1 generative AI SaaS delivery across web engineering, microservice decomposition, model-serving call chains, task queues, logging, monitoring, and platform operations.

The core value was AI productization: connecting model inference, task state, file storage, post-processing, and user experience into usable, scalable, and operable product systems.

Platform Scope

At SenseTime, the AIGC product work included:

1. AI image generation platform supporting proprietary image models, LoRA training, and third-party open-source model integration.
2. Web / H5 / mini-program product surfaces and reusable component systems.
3. Model serving around Stable Diffusion / PyTorch workflows, including parameter formatting, prompt optimization, task queues, model-file storage, and result post-processing.
4. Microservice boundaries for identity, model inference, model-file storage, and task delivery.
5. Engineering governance across containerized deployment, logging, monitoring, service communication, load balancing, and throughput optimization.

Engineering Tradeoffs

The key challenge was bringing generative models into a reliable product system:

• Model call chains needed clear records of parameters, inputs, outputs, latency, and failure reasons.
• Task queues, model services, and file storage had to be separated so long-running inference would not block the web path.
• Prompting, model parameters, post-processing, watermarking, QR overlays, and object storage had to become a reusable pipeline.
• The web experience had to balance ease of use, responsive layouts, and complex task-state feedback.

Platform Value

Established model-serving engineering, long-running task orchestration, cross-platform product experience, and platform operations, extending these patterns into model-call governance, task observability, cost control, and production release workflows for enterprise Agent platforms.