---
title: 'ML Research Scientist at Topos Bio'
canonical: 'https://feeny.ai/job/ml-research-scientist-topos-bio-san-francisco-dwwrzh13p656'
type: 'job'
last_seen: '2026-10-12'
---

# ML Research Scientist at Topos Bio

- **Company:** Topos Bio
- **Location:** San Francisco, CA
- **Employment:** full-time
- **Work type:** onsite
- **Posted:** 2026-09-30
- **Last confirmed live:** 2026-10-12
- **Apply:** https://jobs.ashbyhq.com/toposbio/33fac47a-3c30-439b-a878-c8aa0bd51a32

## Job description

Topos Bio Topos Bio is developing computational methods to drug intrinsically disordered proteins. We're seeking a research scientist to develop and optimize ML methods that bridge simulation, protein modeling, and small molecule design. This role requires both strong machine learning foundations and the ability to work at the interface of computational biology, structural modeling, and chemistry.

## What you will do

- Develop generative models for protein conformational ensembles trained on simulation and experimental data
- Build novel ML architectures for ensemble-based affinity prediction and generative chemistry
- Adapt methods from research literature to disordered protein targets
- Design scalable training pipelines for high-throughput experimentation
- Collaborate with computational and experimental scientists to validate models against biophysical data

## What we are looking for

- Strong foundation in machine learning theory and implementation
- Deep understanding of modern generative models with application to the sciences
- Background in computational sciences involving biology and chemistry

## Preferred

- MS or PhD in machine learning with application to computational science
- Strong publication record
- Experience with molecular simulation and protein modeling

## About Topos Bio

## Company Overview
- **One-liner**: Topos Bio is building an AI-native drug discovery platform that models intrinsically disordered proteins (IDPs) as dynamic ensembles to unlock previously "undruggable" targets in neurodegenerative diseases and cancer.
- **Entity Type**: Private (Seed Stage)
- **Headquarters**: San Francisco, California, United States
- **Founded**: Not publicly available (likely 2024-2025 based on funding and team formation)
- **Founders**: Ryan Zarcone (Co-Founder & CEO), Amir Khosrowshahi (Co-Founder & CTO), Fernando Martinez (Founding Scientist)

## Core Business
- **Primary industry/industries**: Biotechnology, AI-Powered Drug Discovery, Therapeutics
- **Target customers**: B2B (pharmaceutical and biotech partners) and internal pipeline for drug development; enterprise-level research institutions.
- **Mission or purpose statement**: To understand and control biology in motion by modeling intrinsically disordered proteins (IDPs) as dynamic ensembles, unlocking previously "undruggable" targets for diseases like Alzheimer's, Parkinson's, and cancer.

## Products & Services
- **Topos-1 Foundation Model**: An all-atom foundation model trained on a large proprietary dataset of simulation and experimental data. It generates physically realistic conformational ensembles for flexible proteins at speeds that make ensemble-level modeling practical for daily research and drug discovery. (Type: AI Model / Platform)
- **AI-Native Drug Discovery Platform**: An end-to-end platform that uses Topos-1 to design small molecule and biologic therapeutics specifically for intrinsically disordered proteins (IDPs). The platform integrates proprietary data generation, computational modeling, and experimental validation. (Type: Platform / Service)
- **Internal Drug Pipeline**: Topos is developing its own therapeutic candidates targeting IDPs implicated in Alzheimer's disease, Parkinson's disease, and aggressive cancers (e.g., prostate cancer). (Type: Pipeline / Therapeutics)

## Market Standing
- **Valuation/Market Cap**: Not disclosed (Private company)
- **Key Metric**: Total Funding of **$10.5M** (Seed Round, announced January 2026)
- **Notable Investors/Partners**: Threshold (Lead), Neo, and other investors. Partners include **Gladstone Institutes** (neuroscience research) and **Dalriada Drug Discovery** (mass spectrometry and proteomics).
- **Growth Signals**: 
    - Launched its first foundation model, Topos-1, in early 2026.
    - Published a technical report detailing Topos-1's capabilities.
    - Formed a strategic partnership with Dalriada Drug Discovery in March 2026.
    - Headcount of 16 employees with a monthly growth rate of +11.1% (as of mid-2026).
    - Active job posting for a Computational Chemist.

## Competitive Advantages
- **Focus on the "Undruggable"**: Targets intrinsically disordered proteins (IDPs), which represent ~33% of the human proteome and are implicated in major diseases like Alzheimer's and cancer, but are ignored by most drug discovery platforms that rely on static structures.
- **Proprietary Data Moat**: Built a large, proprietary dataset of simulation and experimental data specifically for IDPs, as no public dataset exists for this purpose. This creates a significant barrier to entry for competitors.
- **Ensemble Modeling Approach**: Unlike AlphaFold and other static structure predictors, Topos-1 models the full conformational ensemble of a protein, capturing its motion and dynamics. This is a fundamentally different and more accurate approach for targeting flexible proteins.
- **Integrated Experimental Flywheel**: The company has built in-house experimental capabilities to test and fine-tune its AI models, creating a data flywheel that accelerates compound discovery.

## Strategic Focus
- **Validate the Platform**: Continue to publish technical results and partner with leading research institutions (like Gladstone Institutes) to validate Topos-1's predictive power.
- **Advance Internal Pipeline**: Progress therapeutic candidates for neurodegenerative (Alzheimer's, Parkinson's) and oncology (prostate cancer) targets toward preclinical development.
- **Expand Partnerships**: Leverage the Dalriada partnership to integrate advanced proteomics data, further improving the model's accuracy and scope.
- **Scale the Team**: Grow the headcount, particularly in computational chemistry and research science roles, to accelerate platform and pipeline development.

## Why Work Here
- **Mission-Driven**: The opportunity to work on a fundamental problem in biology—modeling protein dynamics—with the potential to cure some of the most devastating human diseases.
- **Cutting-Edge Science**: Work at the intersection of AI (generative models, foundation models) and biophysics (molecular dynamics, structural biology) on a novel approach that challenges the status quo.
- **Hybrid Workspace**: The company operates with a hybrid model, allowing for a combination of remote and on-site work in San Francisco.
- **Early-Stage Impact**: As a small, well-funded seed-stage company (16 employees), new hires will have significant ownership and influence over the platform and pipeline direction.
- **Strong Founding Team**: Led by experienced founders and scientists from top institutions (Berkeley, Stanford, Intel, Rezo Therapeutics), with a board including notable tech and biotech figures.

## Sources
1. [toposbio.ai](https://toposbio.ai/)
2. [toposbio.ai/about](https://www.toposbio.ai/about)
3. [linkedin.com](https://www.linkedin.com/company/topos-bio)
4. [builtin.com](https://builtin.com/company/topos-bio)
5. [jobs.ashbyhq.com](https://jobs.ashbyhq.com/toposbio)

## Other roles at Topos Bio

- [Computational Chemist](https://feeny.ai/job/computational-chemist-topos-bio-san-francisco-hapthyx8b447) — San Francisco, CA
- [Cellular Biologist](https://feeny.ai/job/cellular-biologist-topos-bio-south-san-francisco-3tdhe9fqcb2s) — South San Francisco, CA