Candidate fit35%75
JD evidence
LLM inference and generative AI
CUDA or Triton kernels
Distributed inference
Candidate proof
Documented ML project
Time-series portfolio evidence
Gaps
Five or more years in performance-critical systems
NVIDIA · Tel Aviv
Senior Software Engineer – LLM Inference
Newly verified LLM-inference systems role covering MoE, SSMs, quantization, CUDA/Triton and distributed deployment.
stretch75opportunity score
Verified openChecked 7 Sept 2026
Hard-constraint gateVerify before applying
- work authorizationIsrael work authorization
Sponsorship is not stated on the official posting.
- years5+ years of performance-critical software engineering
The official NVIDIA requisition explicitly requires five or more years.
Responsibilities
- Optimize LLM and omnimodal inference
- Ship production features in vLLM, SGLang and related runtimes
Requirements
- LLM inference and generative AI
- CUDA or Triton kernels
- Distributed inference
- Five or more years in performance-critical systems
Compensation: not stated
The employer did not state compensation. No estimate is presented as official.
Explainable score · 2026-08-v1
Why this score?
75/100
Career upside20%83
JD evidence
Research-to-production scope
Candidate proof
Research-to-industry portfolio
Gaps
None material
Compensation10%70
JD evidence
Candidate proof
Gaps
Employer compensation not stated
Actionability20%72
JD evidence
Official source captured
Candidate proof
Targeted CV draft available
Gaps
Israel work authorization
5+ years of performance-critical software engineering
Personal factors15%82
JD evidence
Candidate proof
English-working market preference
Gaps
None material
Source confidence 96%Explanation coverage 95%