Start with important questions
Across Virtual Cells, genomic foundation models, and autonomous scientific discovery, we choose questions with lasting value that warrant sustained research.
Our Work
Turn long-term ambition into daily modeling, experimentation, and validation.
Many diseases still lack effective understanding and treatment. We believe that connecting AI, experimental science, and automation in a closed loop can help us understand cell states, validate disease mechanisms, and improve the efficiency of scientific discovery.
We begin with concrete questions. Algorithm, biology, and engineering teams define the path together, then test their judgment through real experiments. Long-term work requires patience, and meaningful progress comes from rigorous research and collaboration every day.
Advance through cross-disciplinary work
Algorithm researchers, biologists, experimental scientists, and automation engineers define problems, break down paths, and share responsibility for results.
Build results that withstand validation
The value of a model must be answered by experiments. We respect evidence, care about reproducibility, and take every unexpected result seriously.
Our Team
A cross-disciplinary team spanning research, engineering, and industry.
Our team works at the intersection of biology, artificial intelligence, mathematics, and drug discovery, combining frontier research with experience bringing technology into real R&D settings.
Academic BackgroundMIT, Tsinghua University, Peking University, University of Chicago, Carnegie Mellon University, Zhejiang University, and others
Industry & Research ExperienceBroad Institute, Harvard Medical School, Sanofi, Arc Institute, XtalPi, Microsoft Research, Alibaba, IBM Research, and others
Open Roles · 09
We are looking for like-minded people.
Whether your field is algorithms, biology, cell experimentation, automation, or organizational development, we value strong fundamentals, independent judgment, candid collaboration, and the ability to carry research through to real results.
AI Research5 Roles
01Senior Algorithm Research ScientistDevelop new mathematical representations and reasoning paradigms for complex living systems from first principles
What You Will Do
- Study the underlying data structures of complex biomolecules and interaction networks, and develop reduced yet high-fidelity mathematical representations.
- Design algorithms that can autonomously generate scientific hypotheses and self-evolve through physical validation feedback, grounded in the philosophy of science.
- Optimize generative AI and large language model training on multimodal biological data to improve reasoning and robustness under complex biophysical constraints.
What We Look For
- PhD in mathematics, theoretical physics, computer science, or a related field, with strong mathematical reasoning and system-level modeling ability.
- Deep expertise in information theory, complex network science, graph learning, or deep generative models.
- Ability to understand the physical and statistical meaning of biological data, with sustained interest in life science and first-principles modeling.
02AIVC Virtual Cell Foundation Model ResearcherUse multi-omic foundation models to understand, predict, and design cell states
What You Will Do
- Build AI+Bio foundation models around multi-omic and multimodal data.
- Develop single-cell perturbation models with stronger generalization to unseen perturbations, cell types, cell lines, and patients.
- Work with experimental teams to create minimal validation loops, using model predictions to guide wet-lab experiments and experimental feedback to improve models.
- Support real applications including drug screening, target discovery, cell reprogramming, cell therapy design, and disease mechanism reasoning.
What We Look For
- Experience with deep generative models, representation learning, sequence modeling, graph models, or foundation model training.
- Proficiency with PyTorch or JAX and the ability to run complete model research and experimentation workflows.
- Understanding of single-cell omics, Perturb-seq, CRISPR screens, drug perturbation, multi-omics, or spatial omics.
- Ability to independently handle data processing, model design, training, evaluation, analysis, and paper-level reproduction.
Preferred Qualifications
- Experience in single-cell foundation models, perturbation or drug response prediction, cell-state modeling, or multi-omic integration.
- Familiarity with scGPT, Geneformer, scVI, CPA, GEARS, STATE, CellFM, or related work.
- Large-scale model training, strong benchmark results, top-tier publications, or a working understanding of biological experiments.
03Self-Evolving Agent Algorithm ResearcherBuild agents that improve through scientific tasks, experimental feedback, and long-term memory
What You Will Do
- Develop multi-agent systems for biomedical discovery, covering hypothesis generation, evidence retrieval, mechanism reasoning, and experimental design.
- Build self-evolving agent frameworks that improve using literature, model predictions, experimental results, and expert evaluation.
- Research long-term memory, task decomposition, tool use, planning, and reflection to improve stability and control in complex research tasks.
- Work across algorithm, biology, experiment, and product teams to bring agents into real scientific workflows.
What We Look For
- Experience with LLM agents, multi-agent systems, tool use, RAG, planning, reflection, memory, or workflow orchestration.
- Strong foundations in machine learning, deep learning, or reinforcement learning, with the ability to read and reproduce frontier research.
- Proficiency in Python and familiarity with PyTorch, JAX, Transformers, LangGraph, LlamaIndex, vLLM, Ray, or related stacks.
- Both research and engineering ability, with a record of defining goals, validating ideas, and iterating in open-ended problems.
Preferred Qualifications
- Research or open-source work in LLM agents, AI Scientist systems, automated science, knowledge graphs, or causal reasoning.
- Understanding of target discovery, disease mechanisms, drug screening, cell therapy, or CRISPR perturbation experiments.
- Background in complex systems engineering, reinforcement learning, MCTS, program synthesis, or neuro-symbolic reasoning.
04Autonomous Scientific Discovery ResearcherRepresent scientific discovery as a system that can be computed, executed, and evaluated
What You Will Do
- Study scientific discovery in the AI era and represent the generation and validation of scientific questions as computable, executable, and evaluable systems.
- Build evaluation frameworks for hypothesis novelty, evidence strength, mechanism completeness, falsifiability, and experimental value.
- Design autonomous systems for biomedical discovery and develop self-evolving Scientific Agent frameworks.
- Continuously validate system capability through automated experiments and real biological questions.
What We Look For
- Experience with LLM agents, multi-agent systems, tool use, RAG, planning, reflection, memory, or workflow orchestration.
- Strong foundations in machine learning, deep learning, or reinforcement learning, with the ability to read and reproduce frontier research.
- Proficiency in Python and familiarity with model and agent engineering stacks.
- Strong interest in scientific discovery, AI for Science, biomedicine, knowledge graphs, causal reasoning, or automated experimentation.
Preferred Qualifications
- Research or open-source experience in AI Scientist systems, automated science, knowledge graphs, or causal reasoning.
- Experience building observable, debuggable, and reviewable complex systems.
- Background in reinforcement learning, MCTS, program synthesis, neuro-symbolic reasoning, formal verification, or top-tier publications.
05Genomic Foundation Model ResearcherModel the path from DNA and RNA sequences to regulation and disease phenotypes
What You Will Do
- Develop genomic foundation models for DNA, RNA, regulatory sequences, and gene embeddings.
- Explore Transformer, Hyena, Mamba, state-space, diffusion, and flow-based architectures for long genomic sequences.
- Build models connecting sequence to expression regulation, chromatin state, transcription-factor binding, variant effects, and disease phenotypes.
- Study long-context genomic modeling, cross-species generalization, and generalization to unseen genes, variants, and cell types.
- Work with biology teams on disease mechanism interpretation, regulatory element design, functional variant prediction, and candidate target analysis.
What We Look For
- Experience with foundation model training methods including Transformers, self-supervised learning, masked modeling, contrastive learning, or generative modeling.
- Expertise in at least one area of genomics, transcriptomics, epigenomics, regulatory elements, transcription factors, eQTLs, or variant effect prediction.
- Proficiency in Python and PyTorch or JAX, with experience in large-scale training, distributed training, or high-performance inference.
- Ability to independently process large biological sequence and multi-omic datasets.
Preferred Qualifications
- Research experience in genomic foundation models, DNA/RNA/protein language models, or single-cell foundation models.
- Familiarity with Enformer, Nucleotide Transformer, DNABERT, Evo, HyenaDNA, Caduceus, Basenji, ESM, or related work.
- Experience in multi-omic integration, large-scale pretraining, top-tier publications, or high-quality open-source contributions.
Biology & Automation3 Roles
06Cell Experiment EngineerTurn precise cell operations into stable, verifiable, and automatable experimental capabilities
What You Will Do
- Perform advanced cell experiments, with emphasis on primary cell culture and directed differentiation and expansion of stem cells.
- Work with automation engineers to decompose passaging, media exchange, resuspension, and other manual operations into executable device actions, then test and debug them.
- Establish rigorous aseptic standards and cell-bank quality systems to maintain stable cell states during scaled automated operation.
What We Look For
- Master’s degree or above in cell biology, bioengineering, or a related field; exceptional practical experience may substitute for the degree requirement.
- Deep expertise in cell culture, with substantial hands-on experience in stem-cell culture and complex directed differentiation.
- Close attention to experimental detail, sensitivity to subtle biological state changes, and consistent execution of high-quality protocols.
07Automation EngineerBuild laboratory automation systems from the ground up to understand experiments, connect devices, and operate at scale
What You Will Do
- Own the architecture, equipment selection, and deployment of a biological automation laboratory, especially for automated cell culture.
- Integrate hardware and software vendors, including secondary development, API integration, and system integration.
- Solve throughput, error-control, and redundancy challenges as systems move from prototype validation to high-throughput operation.
- Work with multi-agent orchestration systems to ensure that high-level commands are executed accurately by physical hardware.
What We Look For
- Bachelor’s degree or above in automation, mechatronics, biomedical engineering, or a related field.
- Experience leading or deeply contributing to biological laboratory automation lines.
- Strong knowledge of system integration for cell culture, liquid handling, and related automation equipment.
- Familiarity with PLCs, robotic arm kinematics, and at least one low-level device communication protocol.
08Senior Scientist, BiologyAdvance computational predictions into preclinical evidence through disease biology and translational validation
What You Will Do
- Lead core programs in oncology, immunology, or metabolism from target validation and mechanism studies through animal-model testing.
- Design rigorous in-vitro and in-vivo validation experiments around disease mechanisms and manage critical preclinical milestones.
- Work with automation teams to convert complex protocols into machine-executable, high-throughput SOPs.
- Partner with algorithm teams to validate biological predictions from computational models.
What We Look For
- PhD in oncology, immunology, metabolic biology, or a related field.
- Three to five or more years of drug R&D experience in pharmaceutical or biotech companies.
- Deep expertise in the pathology of a relevant disease area and extensive hands-on translational medicine experience.
- Openness to new technologies and the ability to work effectively in data-driven, high-throughput validation environments.
People & Operations1 Role
09People & Administration ManagerBuild the organizational capabilities that support scientific innovation and long-term growth
What You Will Do
- Develop medium- and long-term people and administration plans and annual budgets aligned with the company’s three-to-five-year strategy.
- Own the full people function, with an early focus on critical hiring and competitive incentive programs.
- Coordinate administration, finance liaison, government affairs, intellectual property, and day-to-day operations.
- Work with the CEO and management team on company operations and organizational development.
What We Look For
- Bachelor’s degree or above and at least eight years of human-resources experience in high-technology industries.
- Hands-on experience building people and administration functions from zero to one in a startup, plus basic financial knowledge.
- Understanding of AI and biopharma talent markets, with the ability to independently take critical hiring from role definition through successful onboarding.
- Experience implementing incentive and performance systems, with integrity, accountability, results orientation, and strong communication and coordination.
Preferred Qualifications
- End-to-end experience designing and implementing equity incentive or option programs.
- Strong interest in AI and practical experience using it to improve people and administration operations.
Work With Us
If you are interested,
we welcome you to join us.
Send us your CV and the paper, project, or work that best represents you. We also welcome a note about the research questions you care about over the long term. Even if there is no exact role today, we are glad to meet people with strong fundamentals, a serious approach to their work, and sustained interest in this mission.