# Ashe Wang

> 10 X Engineer, Always learning

Location: Sunnyvale, California, United States
Profile: https://flows.cv/ashe

Experienced in large-scale ads infrastructure (targeting, coarse-to-fine ranking, compliance & regulation) and recommendation systems, with a strong foundation in high-throughput data pipelines built using Kafka, Flink, and Spark.

Currently focused on lyrics generation systems, with hands-on experience building and operating LLM workflows for lyrics in production. This includes prompt engineering for creative text, multi-LLM model selection and routing via OpenRouter, and fine-tuning models to improve lyrical coherence, style control, and diversity.

Specialized in lyrics evaluation and observability, designing evaluation pipelines to measure quality, stylistic adherence, repetition, and regression across models. Experienced with DeepEval and custom metrics for tracking quality, latency, cost, and model drift in real-world lyrics generation systems.

Strong systems background with proficiency in C++, Go, and Python, production experience across AWS and GCP, and hands-on experience with PyTorch and large-scale data processing.

## Work Experience
### Staff Software Engineer @ Suno
Jan 2024 – Present
- Shipped major product features end-to-end: Vox Personas (AI voice cloning), Studio             
  (collaborative music editor with versioning), Stems, and Lyrics Mashups—from data modeling       
  through API to frontend                                                                          
  - Built lyrics generation infrastructure: Integrated 10+ LLM providers, designed A/B             
  experimentation framework, implemented content moderation, and iteratively engineered prompts to 
  improve output quality                                                                           
  - Developed audio fingerprinting systems for content identification and neural audio embeddings  
  - Optimized platform for scale: Batch query optimizations, Redis caching strategies, DynamoDB    
  migrations, and hybrid storage architectures to reduce latency and handle 1M+ DAU load           
  - Built evaluation frameworks for AI agents: Designed test suites for infinite loop detection,   
  hallucination detection, and response quality scoring with LLM-as-a-judge; integrated Datadog LLM
   Observability for production monitoring

### Senior Software Engineer @ ByteDance
Jan 2022 – Jan 2024 | Mountain View, California, United States
Ads Infrastructure, Privacy enhancing techniques, ATT restrictions

### Software Engineer @ ByteDance
Jan 2021 – Jan 2022 | Mountain View, California, United States
Ads CSP team 

Ad Infra
-- Monetization Strategy
-- Compliance & Privacy

### Software Engineer @ TikTok
Jan 2021 – Jan 2022 | Mountain View, California, United States
Ads Infrastructure

### Software Development Engineer @ Google
Jan 2018 – Jan 2021 | Sunnyvale
Google Cloud AI

### Quantitative Developer Intern @ AmeriHome Mortgage Company, LLC
Jan 2017 – Jan 2018 | Greater Los Angeles Area

### Research Assistant @ The University of Chicago Booth School of Business
Jan 2017 – Jan 2017
Accelerate algorithm using CUDA

### Research Assistant @ Peking University
Jan 2015 – Jan 2016 | Beijing
Literature reviewed algorithms of Nonnegative Orthogonal matrix Factorization

Implemented  Augmented Lagrangian Algorithm for Nonnegative Orthogonal matrix Factorization using MATLAB and Julia

Improved the algorithm using randomized matrix decomposition

### Business Intelligence Intern @ SAP
Jan 2015 – Jan 2015 | Beijing City, China


## Education
### Statistics
UCLA

### Bachelor of Science (B.S.) in Mathematical Statistics and Probability
Peking University


## Contact & Social
- LinkedIn: https://linkedin.com/in/ashe-wang-999b45132

---
Source: https://flows.cv/ashe
JSON Resume: https://flows.cv/ashe/resume.json
Last updated: 2026-04-01