Team Overview:
LinkedIn’s Core AI organization is dedicated to transforming the professional world through innovative solutions, including advanced models, agents, and AI systems.
Our ‘HALO’ Evaluation Engineering team builds core technology that powers LinkedIn’s model and agent evaluation ecosystem. This is a horizontal and deeply cross-functional team empowering product, linguists, and operations partners to evaluate new AI solutions quickly, efficiently, and at scale.
We’re building a next-generation AI evaluation and optimization platform that makes AI systems measurable, reliable, and continuously improving in production. As AI systems become more autonomous and agentic, evaluation can’t rely on manual labeling and disconnected tools. Our team is creating a unified intelligence layer that connects human feedback, AI judges, synthetic data, training pipelines, and real-time monitoring into a closed-loop improvement engine — defining how AI agents are validated, shipped, and improved at scale.
Team Scope & Future Work
Golden Dataset Generation Tools – Automate creation, labeling, quality control, and versioning of high-quality evaluation datasets from specs and production data.
LLM-as-a-Judge Infrastructure – Build and align large model evaluators to reliably score outputs, reasoning traces, and agent behavior.
Distilled In-House Evaluator Models – Convert large judges into efficient internal models for scalable, low-latency evaluation.
Synthetic Data Generation – Generate controlled, edge-case, and stress-test datasets to expand coverage and robustness.
Observability for AI Agents – Measure hallucinations, tool-use accuracy, reasoning quality, and convergence in real time.
End-to-End Agent Evaluation Framework – Standardize offline benchmarking, regression testing, and production quality monitoring.
Training Signal Pipeline (SFT & RLHF) – Turn evaluation signals and datasets into structured training data for continuous model improvement.
Location:
At LinkedIn, our approach to flexible work is centered on trust and optimized for culture, connection, clarity, and the evolving needs of our business. The work location of this role is hybrid, meaning it will be performed both from home and from a LinkedIn office on select days, as determined by the business needs of the team.
This role will be based in Sunnyvale, CA.
Responsibilities:
Provide direct leadership for a team of 5-7 engineers.
Work closely with various other teams across LinkedIn to align on initiatives, leveraging existing solutions, and ensure optimal collaboration.
Participate in and facilitate technical and design discussions
Contribute to the hands-on execution of technical solutions (coding, design, etc.).
Leverage best practices & processes to ensure productivity of the team and drive faster iterations
Attract world class talent and provide technical guidance, career development, and mentoring to team members.