The posting
The Role:
Generate is seeking a creative and rigorous scientist to develop computational technologies and lead applied platform research that expand what protein therapeutics can do. You will develop reusable methods, scoring and evaluation tools, model benchmarks, agentic systems, and scientific workflows for protein design, engineering, and optimization. Your work will span creating new proteins, modifying existing molecules, and integrating molecular components to unlock new therapeutic applications.
At Generate, model development, protein design, high-throughput experimentation, structural characterization, and therapeutic development operate within one organization. This role offers the opportunity not only to advance computational methods, but to see them shape what proteins are built and tested, learn from rich in-house experimental data, and contribute to medicines in our pipeline. We are looking for someone who is motivated by both technical innovation and its application to drug development.
This role combines technology development with applied platform research. You will identify recurring scientific needs, lead research efforts, and turn successful approaches into durable capabilities with clear interfaces and validated performance. Working with program teams, biologists, and experimental scientists, you will test and refine these capabilities in relevant biological settings, establish proof of concept for new molecular properties and functions, and combine validated approaches to enable new therapeutic applications and pipeline programs. You will lead the computational research needed to establish feasibility, resolve technical risks, and make successful approaches reusable across targets and programs.
Here's how you will contribute:
- Lead technology development and applied platform research. Identify high-value capability gaps and molecular or biological challenges, formulate testable hypotheses, and define research plans, milestones, and decision criteria. Drive projects from scientific question through computational implementation and experimental proof of concept to validated, reusable capabilities.
- Build and evaluate reusable protein engineering capabilities. Develop computational methods for protein design, optimization, and interaction modeling using generative, predictive, and biophysical approaches. Create scores and benchmarks to assess sequences, structures, biological functions, and developability. Define intended uses, interfaces, assumptions, and success criteria across targets, molecular formats, and therapeutic applications. Build reproducible datasets and evaluation protocols that reveal strengths and failure modes, quantify uncertainty, and guide appropriate use.
- Engineer and validate new molecular capabilities. Partner with assay, structural, and experimental scientists to design informative molecular libraries and iterative design–build–test–learn studies. Demonstrate control of properties such as binding specificity, conditional activity, molecular assembly, stability, or developability. Analyze experimental readouts, distinguish signal from noise, diagnose failures, and use the results to improve methods and establish where they generalize.
- Scale scientific capabilities. Partner with machine learning and engineering to translate platform and application needs into model objectives, evaluation criteria, and data requirements. Stress-test new models and work with engineers to turn successful research into reliable, maintainable software.
- Build agentic systems for therapeutic protein design. Develop agentic systems and multi-agent workflows that connect scientific information, generative and predictive models, internal data, analysis tools, and experimental results to formulate hypotheses, propose designs, evaluate evidence, and recommend informative next steps. Establish mechanisms for enabling human oversight, traceability, and rigorous measures of reliability.
- Enable new therapeutic applications. Work with application teams to combine validated capabilities into approaches that address consequential molecular or biological challenges. Establish technical feasibility in therapeutically relevant settings, assess performance against appropriate references, and test whether an approach can support multiple targets or programs. Generate evidence that informs advancement or discontinuation, target selection, and nomination of differentiated therapeutic programs.
- Transfer and disseminate successful approaches. Turn research outcomes into documented methods and reference workflows that other teams can use and interpret reliably. Support adoption by application and program teams, and translate lessons across projects into improvements that benefit the broader platform.
The Ideal Candidate will have:
- Strong scientific and quantitative foundations, with practical fluency in modern machine learning and computational methods relevant to protein science, molecular modeling, or biological design.
- A record of translating research ideas into reusable computational methods, models, software, or workflows, and demonstrating their value on biological problems.
- Experience designing rigorous benchmarks or evaluation frameworks, including dataset construction, baselines, metric validation, uncertainty, confounding, and failure analysis.
- Experience tackling applied biological or therapeutic problems with teams of ML researchers, engineers, protein scientists, biologists, and experimental scientists, including framing hypotheses, defining informative studies, and using results to guide decisions.
- Strong hands-on computational skills, including Python, modern ML frameworks such as PyTorch or JAX, and experience contributing reliable code in a shared environment.
- Strong understanding of protein structure, sequence–structure–function relationships, molecular interactions, and biophysics, with the ability to reason critically about computational predictions and experimental evidence. Relevant experience may include protein or antibody engineering, molecular modeling, structure-guided optimization, generative design, or related approaches.
- Strongly valued: Experience building LLM-enabled agents, scientific copilots, or multi-agent systems for therapeutic design or scientific hypothesis generation and testing, including tool use, planning, evaluation loops, human oversight, and assessment of reliability.
- A PhD in a relevant scientific or engineering discipline, with at least three years of industry experience following completion of the PhD, and a demonstrated ability to lead consequential technical work with substantial independence.
- Background: Candidates may come from computational protein design or engineering, computational biology, machine learning, biophysics, or adjacent quantitative disciplines. The strongest candidates will combine scientific depth with a platform mindset and a desire to see their work tested experimentally and used to create new therapeutics.
Who Will Love This Job:
You bring scientific creativity, strong technical judgment, and disciplined execution to open-ended problems. You value reliability, reuse, and clear evidence; collaborate with a low-ego approach; and move comfortably between original research, robust implementation, and experimental learning. You can distinguish an exploratory result from a validated capability, make clear recommendations from incomplete evidence, and connect technical progress to therapeutic opportunities.
Location:
This role is based in our Somerville, Massachusetts office with flexibility for hybrid work.
#LI-HM1
About Generate Biomedicines
We are a clinical-stage generative biology company pioneering the AI revolution in drug design and development. We are advancing a new approach to drug creation—one grounded in the ability to design proteins with defined biological intent. By integrating machine learning with large-scale experimentation, this approach aims to reduce the uncertainty, time, and cost associated with developing protein-based medicines.
Founded in 2018, we are advancing a growing pipeline of clinical and preclinical programs across multiple disease areas and protein modalities. By unifying computational design and clinical development within a single operating model, we translate this approach into clinical-stage programs and are leading a shift from traditional drug discovery toward systematic drug generation.
At Generate:Biomedicines, we collaborate across disciplines in new ways to invent and innovate. We bring diverse perspectives to a shared goal of delivering better medicines to patients in need, faster, guided by our values and leadership behaviors.
Generate:Biomedicines is committed to equal employment opportunity regardless of race, color, ancestry, religion, sex, national origin, sexual orientation, age, citizenship, marital status, disability, gender identity or Veteran status.
Recruitment & Staffing Agencies: Generate:Biomedicines does not accept unsolicited resumes from any source other than candidates. The submission of unsolicited resumes by recruitment or staffing agencies to Generate:Biomedicines or its employees is strictly prohibited unless contacted directly by the Company’s internal Talent Acquisition team. Any resume submitted by an agency in the absence of a signed agreement will automatically become the property of Generate:Biomedicines and the Company will not owe any referral or other fees with respect thereto.
Compensation: The base salary range provided reflects our current estimate of what we anticipate paying for this position. Your actual base salary will be based on several factors, including job-related skills, experience, internal equity, relevant education or training, and market dynamics. In addition, you will be eligible for an annual bonus, equity compensation, and a competitive benefits package.
Per Year Salary Range
$172,000—$241,000 USD
