Customer Success Engineer
AI inference cloud providing OpenAI-compatible APIs, private deployments, and GPU infrastructure for production AI workloads.
Funding history
About DeepInfra
DeepInfra operates an AI inference cloud for running LLMs, vision, embeddings, image/video generation, speech, and other machine-learning models at scale, including private GPU deployments and GPU rental.
Skills
Candidate Availability
Required and preferred rules are kept separate and reflect the wording in the original posting.
About the Role
You will own technical customer issues from triage through resolution across inference workloads. You will investigate logs, metrics, dashboards, API integrations, billing, quotas, and capacity requests. You will help customers optimize platform usage and turn recurring problems into documentation, runbooks, bugs, and product improvements.
Requirements
- 2+ years of experience in customer success engineering, support engineering, solutions engineering, technical account management, or similar roles, or software engineering experience with interest in customer work
- Computer Science or Engineering degree or equivalent hands-on background
- Knowledge of HTTP, REST APIs, status codes, headers, streaming responses, authentication, and integration code
- Python or a similar programming language
- Command-line, log, metric, and dashboard experience
- Experience with Grafana, Prometheus, Loki, or equivalent tools
- Written technical communication skills
- Ability to work independently and prioritize a support queue
Responsibilities
- Own customer technical issues from triage through resolution
- Debug production inference issues using logs, metrics, and dashboards
- Analyze traffic, concurrency, and token throughput for capacity requests
- Resolve billing, usage, and quota questions
- Advise customers on platform usage and model deployment choices
- Plan capacity and dedicated deployments
- File and drive bugs and feature requests
- Create documentation, runbooks, and product improvements for recurring issues
- Surface customer patterns to Product and Engineering