AfterQuery
Research Scientist - Post Training
$210K–$450K
San Francisco · On-site · Full-time · Posted Apr 14, 2026
About the job
Designs and runs post-training experiments to show how AfterQuery’s datasets affect foundation-model capabilities, generalization, and alignment.
What you'll do
- Run controlled SFT and RL experiments to measure dataset impact on model performance.
- Help build public evaluations and new data types; publish research, blog posts, and technical reports.
- Use results to work with internal SPLs on improving data quality.
What they ask for
- Strong familiarity with LLM training and evaluation methodologies.
- Ability to design lightweight experiments, move fast, and extract actionable insights from messy results.
- Comfort working across domains, including finance, software engineering, and policy.
- Bias toward building over theorizing.
Nice to have
- Undergraduate or master’s research experience (without a PhD).
- Strong interest in how data structure, selection, and quality drive model behavior.
Pay
$210K – $450K (base salary); $250k-450k (total compensation)
Tools
SFT, RL
AI focus
Conduct SFT and RL-based post-training research to evaluate how datasets influence model behavior, capability, generalization, and alignment.
Model-written summary · the original posting is authoritative