About the position
You will be responsible for designing, implementing, and operating the core backend systems for Cotomo and our upcoming new products.
TechStack
Python, Rust, TypeScript, WebSocket, WebRTC, ElasticSearch, PostgreSQL, GCP, Azure, AWS, Unity, Weights & Biases, NVIDIA Triton, vllm, pytorch, transformers, deepspeed, Dataform, BigQuery, Sentry, Slack, Github
Responsibilities
- Design and implement efficient and highly available infrastructure to support high-volume requests
- Build backend systems that integrate with various AI models including STT, LLM, and TTS.
- Develop streaming systems to enable real-time, high-quality voice conversations
- Construct large-scale data analysis platforms
- Optimize performance and improve scalability of our systems
Requirements
- 3+ years of experience in designing, implementing, and operating backend systems
- Proficiency with relational databases (PostgreSQL/MySQL etc.) and NoSQL databases
- Basic knowledge of real-time communication technologies such as WebRTC and WebSocket
- Experience in operating systems on cloud platforms (AWS, GCP, Azure, etc.)
- Experience in developing applications using RAG - personal projects are acceptable
- Experience in building and improving CI/CD pipelines
- Conversational Japanese ability
Nice to haves
While not specifically required, tell us if you have any of the following.
- Enthusiasm for learning and applying new technologies to product development
- Ability to think from a user experience perspective and creatively solve technical challengesValues teamwork and can communicate openly
- Experience working in early-stage startups (within a few years of founding)
- Experience in operating machine learning models in production environments
- Experience in building and maintaining home server environments
- Experience in training and fine-tuning deep learning models such as LLMs
- Knowledge or experience in speech recognition and natural language processing
Compensation
Starting from 7.5 million JPY annually, with performance-based stock options.
About Starley
Starley designs new relationships between people and AI by developing products that fit into everyday life.
They are developing "Cotomo", an audio-based AI conversation app, which users can use to talk about everyday life or even more personal topics.
Starley aims to redefine how AI and people communicate and to provide the world with new experiences.
Get Job Alerts
Sign up for our newsletter to get hand-picked tech jobs in Japan β straight to your inbox.