Wesley Wei Qian
Experience
Founding Team | Osmo
Sep 2022 - Present
- First research engineer; now VP of Engineering & Research. Lead a 10+ person team building AI products that digitize the sense of smell — from prototype to production.
- LLMs & Agentic Systems: Built a multi-modal RAG system that turns text and image briefs into physical fragrance samples ($100K+ in sales within 3 months). Fine-tuned LLMs (GPT, LLaMA) with SFT and RLHF using proprietary perfumery tools to match expert preferences.
- 0-1 ML Stack: Built the post-Google spinout ML stack using PyTorch, GCP, and W&B. Engineered and productionized GNNs and an end-to-end odor intensity pipeline from data collection to deployment that increased discovery throughput by 10×, enabling partnerships with a Top 4 fragrance house.
- AI for Science: Led a team of ML researchers and analytical chemists to "teleport" a physical plum scent — translating GC-MS chemical signatures into digital profiles for remote reconstruction.
- Leadership & Product Impact: Built Studio, an AI-agentic fragrance creation platform that cut formulation cycles from weeks to hours. Secured a multi-million dollar CPG partnership through AI-driven cost and safety improvements.
Student Researcher | Google
May 2018 - Sep 2022
- Olfactory ML: Built receptor binding and metabolic activity models that led to one utility patent, four publications featured in major news outlets, and catalyzed Osmo's $60M spin-out.
- Genomics & Structural Variants: Developed a new approach to structural variant calling with ML-based filtering; engineered a ~100× faster read-alignment method, filed as a utility patent.
- Generative Cell Images: Developed a GAN to correct batch effects in high-content cell imaging; contributed to Google's TF-GAN library and published in Bioinformatics.
Intern | DeepMind
Sep 2021 - Dec 2021
- AlphaFold Team: Worked on protein-folding research using JAX and the AlphaFold2 codebase, developing representation learning methods for structure prediction tasks.
Software Engineering Intern | Uber
Summer 2016 & 2017
- [2017] Sensor & Machine Learning: Built a CRF variant to infer Uber Eats delivery events from mobile sensor data. Won 1st place at Uber's internal ML poster session.
- [2016] Mobile Dev. & Engineering: Developed a web-based forensics tool for mobile UI testing that synchronized logs with video timestamps, reducing internal developer debugging time by ~50%.
Education
2017 - 2022
Doctor of Philosophy in Computer Science
Dissertation: Machine learning for drug discovery and beyond (advisor: Jian Peng)
GPA: 4.00 / 4.00 | University Fellowship | Richard T. Cheng Endowed Fellowship
2013 - 2017
Bachelor of Science in Computer Science and Neuroscience
GPA: 3.96 / 4.00 | Summa Cum Laude | Phi Beta Kappa (Junior) | Schiff Fellowship
Recent & Selected Publications (*equal contribution)
New York Smells: A Large Multimodal Dataset for Olfaction
arXiv (2025)
A deep learning and digital archaeology approach for mosquito repellent discovery
Chemical Science (2025)
A Principal Odor Map Unifies Diverse Tasks in Human Olfactory Perception
Science (2023)
Metabolic activity organizes olfactory representations
eLife (2023)
3D Equivariant Diffusion for Target-Aware Molecule Generation and Affinity Prediction
ICLR (2023)
Energy-Inspired Molecular Conformation Optimization
ICLR (2022)
Integrating Deep Neural Networks and Symbolic Inference for Organic Reactivity Prediction
ACS National Meeting (2021)
Batch Equalization with a Generative Adversarial Network
Bioinformatics (2020)
Other Publication
Pervasive mislocalization of pathogenic coding variants underlying human disorders
Cell (2024)
A central chaperone-like role for 14-3-3 proteins in human cells
Molecular Cell (2023)
ECNet is an evolutionary context-integrated deep learning framework for protein engineering
Nature Communication (2021)
Comprehensive interactome profiling of the human Hsp70 network highlights functional differentiation of J domains
Molecular Cell (2021)
Evaluating Attribution for Graph Neural Networks
NeurIPS (2020)
Evolutionary context-integrated deep sequence modeling for protein engineering
RECOMB (2020)
Services
- Reviewer for Nature Machine Intelligence (2025), ICML (2024), ICLR (2024), NeurIPS (2023), LoG (2022 & 2023), RECOMB (2021), ISMB (2019 & 2020)
- Program Committee for ICML - ML Interpretability for Scientific Discovery Workshop 2020