[Skip to main content](#main-content)

[OpenNash Jobs](https://jobs.opennash.com/)
[Hiring report](/reports/hiring/) · [All S.F. Holding jobs](https://jobs.opennash.com/companies/sf-holding/) · [Search jobs](/search/)

S.F. Holding · 深圳市

# 大模型算法工程师（多模态）

深圳市 

Listed on S.F. Holding’s official careers site. Posted January 16, 2026. Check that it is still open and apply there.
[Apply on S.F. Holding’s careers site →](https://crs-pub.sf-express.com/recruitcenter/#/positionDetail/110619?jobId=67574&role=3)

[OpenNash Jobs](https://jobs.opennash.com/) / [S.F. Holding](https://jobs.opennash.com/companies/sf-holding/) / 大模型算法工程师（多模态）

## Job description

### 职责

* 针对物流垂域场景进行深入的视觉研发，主要是面向小样本/零样本场景目标的检测 * 负责将大模型能力蒸馏到轻量化学生模型（YOLO 系），落地在边缘端，达成精度-延迟-功耗平衡 * 搭建检测训练与评估流水线：数据清洗/去重、标注规范、数据集版本化，指标包含 mAP、召回率 * 构建半监督/弱监督训练链路，并与业务无标数据结合，提升域内泛化 任职要求：* 计算机/自动化/电子等相关专业，研究生以上学历；2年及以上视觉/检测方向经验（优秀者年限可放宽） * 扎实的 CV/ML 基础：检测框架与常用损失（IoU/DFL/Focal/Varifocal）、数据增强、长尾与不平衡处理 * 熟练使用 PyTorch，掌握至少一种检测框架：MMDetection/Ultralytics YOLO/PaddleDetection/RT-DETR * 有小样本/零样本或半监督/弱监督检测实践，能将开放词表模型（GroundingDINO、OWL-ViT等）用于高召回伪标签与快速冷启动

## At a glance
- Posted by the employer: Jan 16, 2026
- First seen by OpenNash: Oct 12, 2026
- Last checked: Oct 19, 2026
- Listed for: 276 days
- Location: 深圳市[Apply on S.F. Holding’s careers site →](https://crs-pub.sf-express.com/recruitcenter/#/positionDetail/110619?jobId=67574&role=3)

## More jobs at S.F. Holding

### [1通信高科技行业解决方案高级专员](https://jobs.opennash.com/companies/sf-holding/jobs/ba50f13579c91993ce58/)

深圳市

### [22通信高科技解决方案高级专员](https://jobs.opennash.com/companies/sf-holding/jobs/d319e24168eb5661ea1e/)

深圳市

### [23通信高科技解决方案高级专员](https://jobs.opennash.com/companies/sf-holding/jobs/a57eb8257ee302957f85/)

深圳市

### [24通信高科技解决方案高级专员](https://jobs.opennash.com/companies/sf-holding/jobs/f80d35a3b12f1b034594/)

深圳市

### [26通信高科技解决方案高级专员](https://jobs.opennash.com/companies/sf-holding/jobs/146d9458337690fc6f43/)

深圳市

## Similar jobs at other companies

### [大模型算法工程师](https://jobs.opennash.com/companies/ping-an-insurance/jobs/40993513cf79f2efb917/)

Ping An Insurance Group · 深圳市

### [算法工程师-多模态](https://jobs.opennash.com/companies/midea-group/jobs/d0fcc6cd74fa498fcea5/)

Midea Group · 上海市

### [算法工程师-多模态](https://jobs.opennash.com/companies/midea-group/jobs/d848e8ebba6758854125/)

Midea Group · 佛山市 +1 more

### [AI策略产品经理-多模态](https://jobs.opennash.com/companies/xiaomi/jobs/ad827990f617fbfe8f3e/)

Xiaomi · 北京

### [微信小程序-大模型算法工程师-智能体方向](https://jobs.opennash.com/companies/tencent-holdings/jobs/1d10ddb9c87f6e704785/)

Tencent Holdings · Guangzhou

OpenNash lists jobs from employers’ own careers sites and does not handle applications.
Dates come from the employer where stated; “first seen” and “last checked” are our own observations.

OpenNash Jobs · Official listings with dates and links to the employer. [Weekly hiring report](/reports/hiring/) · [Methodology](/methodology/)