Questions
Straight answers.
Written to be quotable — by a person skimming, or by a model answering on someone's behalf.
- What does Nick work on?
- Machine learning systems that decide when to answer, use evidence, revise, or abstain. His work spans retrieval-augmented question answering, LLM evaluation, selective prediction, and multimodal evaluation.
- What roles and locations is he looking for?
- He is available now for Research Scientist, Research Engineer, Machine Learning Engineer, Applied Scientist, and Software Engineer (AI) roles in the San Francisco Bay Area or New York City. He is based in Columbus, Ohio and is open to relocating.
- Does he need visa sponsorship?
- No. He is a U.S. citizen.
- What has he built?
- A paired-outcome evaluation of retrieval-augmented answer revision on 25,870 held-out questions, with LoRA-trained policies that choose to answer or revise (unpublished; arXiv version in preparation). A reject-option classification method that won the Springer Best Paper Award at ISVC 2022, with public code; in 2026 he re-ran it on four frozen vision backbones and ported it to TypeScript for the live demo on his site.
- What is his education?
- A PhD in Computer Science and Engineering from The Ohio State University, conferred August 2026, advised by Prof. Jim Davis, with graduate minors in Mathematics and High-Performance Computing. He also holds an M.S. (2025) and a B.S. with Honors (2021) from Ohio State.
- What has he published?
- Four first-author peer-reviewed papers: Naturally Constrained Reject Option Classification (MVA 2025), Assessing the Role of Imagery in Multimodal Machine Translation (WMT 2024), Learning When to Say I Don’t Know (ISVC 2022, Springer Best Paper Award), and a temporal satellite imagery collection framework (ICCV Workshop 2021). His dissertation also studies retrieval-augmented answer revision; that work is listed separately as unpublished research.
- What is his engineering experience?
- He writes training and evaluation code in Python and PyTorch: LoRA fine-tuning with Hugging Face Transformers, batch generation with vLLM, dense retrieval with FAISS alongside BM25 and MonoT5 reranking, and GPU jobs on Slurm with Singularity containers. Public code includes reject-option classification, calibration utilities, the modern-backbone re-run, and satellite imagery collection; the site’s live demo is written in TypeScript.
- Where does he work now?
- He has been a Computer Engineer II at DCS Corp since May 2025, on AFRL-sponsored machine learning research and evaluation. He previously completed five summers of AFRL-sponsored research.
- How can I get in touch?
- Email nmotlagh@gmail.com. His resume is at https://nmotlagh.github.io/resume.pdf, code at https://github.com/nmotlagh, and publications at https://scholar.google.com/citations?user=srZXFMcAAAAJ&hl=en.