Quelle:
Für dieses Angebot liegt keine vollständige Übersetzung in der gewählten Sprache vor. Unten sehen Sie den Originalinhalt.
Member of Technical Staff, LLM & Production
Hark
San Jose, Vereinigte Staaten
Publication date: 02/11/2026
Hark is an artificial intelligence company building advanced, personalized intelligence that is proactive, multimodal, and capable of interacting with the world through speech, text, vision, and persistent memory. The Omni team is building the next generation of AI experiences beyond text, enabling models to understand and generate content across text, audio, and vision. The role is a hands-on, individual-contributor position for an engineer who sits at the intersection of model research and production systems.
Responsibilities
Build and own end-to-end evaluation suites for LLM capabilities, including task specs, grading logic, and dashboards; Run the eval harness as a production service (sandboxed execution, tool calling, multi-turn and agentic tasks, deterministic replay, caching); Own the data pipeline from source to training, including collection, filtering, dedup, labeling workflows, synthetic data generation, and quality gates; Build RL training environments (reward functions, simulators, verifiers, task generators) and keep them stable across many workers; Work side by side with researchers to turn hypotheses into runnable evals or environments and interpret results; Pull capabilities out of the product layer and into the model, building evals and environments that teach the model to perform those tasks natively; Measure eval regressions, environment health, throughput, and cost; Raise the bar for research code with tests, versioned datasets and evals, and clean boundaries between training, inference, and evaluationRequirements
3+ years in software engineering, including 1-2 years hands-on with LLM training, evaluation, or data infrastructure; Deep Python fluency and real distributed-systems experience; Experience building evaluation infrastructure (designing tasks, graders, and the plumbing around them); Working knowledge of SFT and RL for language models; Production instincts: writing tests, insisting on reproducibility, and treating research tooling as something others depend on; Comfortable in ambiguity next to researchers, able to make half-formed ideas measurableNice to have
Agentic evaluation experience (tool use, code execution, sandbox environments, multi-turn tasks); Hands-on with training frameworks and inference stacks (e.g., PyTorch, JAX, vLLM, SGLang, or in-house equivalents); Experience running data-labeling operations or collecting human preferences at scale; Contributions to open-source evaluation frameworks or public benchmarks; Previous work on verifiers, code sandboxes, or simulation environmentsHow to apply
Apply online: https://www.jobs.pl/en/oferta/103409206/aplikuj
JOBS.pl · https://www.jobs.pl/de/member-of-technical-staff-llm-production-oferta-103409206





