Loading jobs

refreshed daily

FuriosaAI

Software Engineer, LLM Performance & Evaluation

New roles land in Following.

SeoulPosted 24 days ago

Apply on FuriosaAI's career page

Market pay check

Matched to Computer systems and software professionals

Source: Ministry of Employment and Labor wage information system (wage.go.kr), wage distribution by occupation.

What you'd do

Design benchmarks isolating inference features like batching,prefix caching,and speculative decoding.
Measure TTFT,latency distributions,throughput,memory use,and accelerator utilization against baselines.
Own model accuracy evaluation selecting datasets and comparing NPU results to trusted references.

What they want

Demonstrate strong Python skills building maintainable automation,data pipelines,or engineering tools.
Measure and analyze ML inference or complex systems performance including regression investigation.
Understand transformer LLM inference including prefill,decode,batching,and KV-cache behavior.

Nice to have

Use LLM inference frameworks such as vLLM,SGLang,or TensorRT-LLM for benchmarking.
Build evaluation suites with lm-evaluation-harness including prompt and scoring validation.
Know quantization,mixed precision,speculative decoding,and distributed inference trade-offs.

Seoulstart's read

Visa tier (likely): E-7 specialty / professional (likely)
Korean language: Korean not required
English friendliness: High: JD in English, foreign-friendly signals
Context: Korean AI accelerator startup with Silicon Valley ties; hybrid Seoul role benchmarking LLM inference on Furiosa NPUs.

This is Seoulstart's analysis of the public job description, not the employer's stated policy. Verify visa sponsorship, language requirements, and remote allowances directly with the recruiter.

Open posting

Related Seoulstart guides

Similar roles on Seoulstart

Browse all English-friendly jobs in Korea