Adaptive AI Scientist: LLM Evaluation & Routing

Bitdeer Technologies Group; Austin (TX); On-site; USD 140,000 - 210,000; Full time

Job summary

Bitdeer Technologies Group in Austin, TX seeks a senior AI/ML engineer to join Bitdeer AI Lab. You will build evaluation systems for agentic inference, extend LLM evaluation pipelines, and prototype routing strategies.

You will collaborate with MaaS and platform teams to deploy production-ready tooling with reliability considerations. You will work on research prototypes, cost and latency analysis, and contribute to advancing evaluation methodologies for multi-model AI systems in a fast-growing

Qualifications

• Experience in designing or operating evaluation pipelines for LLMs or agents.
• Strong Python programming and PyTorch experience.
• Hands-on familiarity with multi-model APIs and agent harnesses.
• Ability to analyze per-request and per-trajectory quality, cost, latency.
• Publications or substantial open-source contributions welcome.

Responsibilities

• Build evaluation and decision systems for agentic inference and measure trajectory-level performance.
• Own and extend LLM evaluation pipeline and represent task suites.
• Research routing strategies: rule-based baselines, cascades, stage-aware routing, uncertainty-aware selection.
• Collaborate with MaaS and platform engineering on production integration and reliability.
• Own evaluation methodology, routing policy, and research prototypes; production gateway reliability is a shared responsibility.

Skills

Python programming

PyTorch experience

LLM evaluation

Agentic systems

Evaluation pipelines

Tool-using agents

Cost & latency analysis

Experiment design

Online learning

Multi-model APIs

Education

Bachelor’s, Master’s, or PhD in Computer Science / ML / Statistics / Electrical Engineering or related field

Tools

PyTorch

vLLM

SGLang

Data tooling

Job description

Bitdeer Technologies Group in Austin, TX seeks a senior AI/ML engineer to join Bitdeer AI Lab. You will build evaluation systems for agentic inference, extend LLM evaluation pipelines, and prototype routing strategies.

You will collaborate with MaaS and platform teams to deploy production-ready tooling with reliability considerations. You will work on research prototypes, cost and latency analysis, and contribute to advancing evaluation methodologies for multi-model AI systems in a fast-growing