Tech Lead, Applied AI - SearchWe9re hiring Engineering Leads to lead teams within our Applied AI organization.
Applied AI builds the systems that directly determine how fast frontier models improve: evaluation infrastructure, rubric and reward-model pipelines, annotation products, and emerging multimodal capabilities like screen-capture-grounded agent training. This isn9t generic ML infra - it9s the layer where human expertise gets translated into signal that labs can actually train on. You9ll lead a team of engineers, partner closely with product, research, and operations, and help define how Mercor stays at the frontier as the industry shifts from static data labeling to dynamic, verifiable training environments.
You9ll also build direct partnerships with leading AI labs and enterprises, and contribute hands-on to improving frontier model quality through data, evaluation, and systems - work that sits squarely in the areas the field agrees matter most right now: RL with verifiable rewards, rubric-based reward modeling for subjective and agentic domains, and the pipelines that turn raw human demonstrations into training-ready environments.
This role requires high ownership and abstraction: setting direction, driving outcomes, and staying hands-on while leading.
What You9ll Work OnBuild and scale teams
- Coach and develop high-potential engineers
- Establish strong ownership, culture, and execution standards
- Shape the team: define hiring processes, establish engineering practices, and scale a high-performing team from the ground up
Drive Applied AI systems and outcomes
- Own the methodology behind evaluation systems that benchmark and improve model performance - from rubric design and LLM-judge calibration to reward-hacking mitigation
- Lead development of pipelines that convert AI-moderated interviews and human demonstrations into skills, rubrics, and post-training data at scale
- Drive improvements in data quality, operational efficiency, system stability, and scalability across human-in-the-loop workflows
- Push capabilities up the verifiability spectrum - from cleanly-graded tasks toward the harder, subjective, and agentic work where domain expertise and rubric quality become the bottleneck
Own technical direction
- Lead system design for complex AI systems, including multimodal and agentic evaluation pipelines
- Stay close to the technical work and guide engineers through genuinely ambiguous, research-adjacent problems
- Translate emerging research (RLVR, rubric-as-reward, agent grounding from human demonstrations) into concrete engineering roadmaps
Operate across a broad scope
- Partner with product, research, operations, and external AI labs on custom evaluation and training-data engagements
- Work across systems, data insights, and lab-specific partnerships to directly move model quality metrics
Shape the organization
- Introduce lightweight processes as the organization scales
- Hire and develop strong engineering talent
- Help define what engineering management looks like at Mercor
What We9re Looking For- 6-10 years in engineering; 2-3+ years leading teams
- Strong background in building scalable systems
- Comfort with core ML/LLM concepts - post-training, evaluation metrics, reward modeling - deep enough to partner with researchers, not just consume their output
- Ability to lead in ambiguous, fast-moving environments with sound technical judgment
- Proven ability to coach engineers and drive execution
- Strong ownership mindset with pragmatic decision-making
