2026 Summer Intern - Computational Sciences Center of Excellence - Agentic tool for Optical Pooled Screens

2026 Summer Intern - Computational Sciences Center of Excellence - Agentic tool for Optical Pooled Screens

Department Summary

A healthier future. It’s what drives us to innovate. To continuously advance science and ensure everyone has access to the healthcare they need today and for generations to come. Creating a world where we all have more time with the people we love. That’s what makes us Roche. 

Advances in AI, data, and computational sciences are transforming drug discovery and development. Roche’s Research and Early Development organisations at Genentech (gRED) and Pharma (pRED) have demonstrated how these technologies accelerate R&D, leveraging data and novel computational models to drive impact. Seamless data sharing and access to models across gRED and pRED are essential to maximising these opportunities. The new computational sciences Center of Excellence (CoE) is a strategic, unified group whose goal is to harness this transformative power of data and Artificial Intelligence (AI) to assist our scientists in both pRED and gRED to deliver more innovative and transformative medicines for patients worldwide.

Within the CoE organisation, the Data and Digital Catalyst organisation drives the modernisation of our computational and data ecosystems and integration of digital technologies across Research and Early Development to enable our stakeholders, power data-driven science and accelerate decision-making.

This internship position is located in South San Francisco, on-site.

The Opportunity

The XY and Li Labs are seeking a PhD intern with deep expertise in AI Agent architectures to advance our capabilities in Optical Pooled Screens (OPS). Our teams have developed high-performance internal computational frameworks to process terabytes of OPS data. However, optimal execution of these pipelines currently relies on complex, manual configuration by domain experts.

We are looking for a researcher to build an autonomous agentic system capable of driving these internal tools. The goal is to move from "human-in-the-loop" operation to a fully autonomous reasoning engine that can optimize data processing strategies on the fly.

Key Responsibilities:

• Architecture: Design a modular Python-based agent capable of "tool use" (invoking internal binary executables and APIs).

• API & Documentation: Build a production-grade interface for the agent. You must prioritize clear, standard communication protocols to ensure the agent can be easily integrated into broader multi-agent workflows.

• Benchmarking: Establish a rigorous evaluation framework to quantify the agent's ability to converge on optimal processing configurations compared to human experts.

Project Scope

• Autonomous Pipeline Optimization: Design an agent capable of utilizing our internal OPS CLI and Python tools. The agent must autonomously test hypotheses on raw image data to determine optimal parameters for image stitching, registration, and feature extraction.

• Agentic Reasoning & Recovery: Implement a robust reasoning loop (e.g., Plan-and-Solve, ReAct) that can interpret structured quality control logs. The system should effectively diagnose failure modes—distinguishing between data anomalies and configuration errors—and self-correct without human intervention.

• Interoperable System Design: A critical requirement is the ability of this agent to exist within a larger automated ecosystem. You will design and implement a well-documented, standard API (REST/OpenAPI) for the agent. This interface must allow third-party orchestrators or other specialized agents to communicate with your system, submit jobs, and negotiate parameters programmatically.

Program Highlights

• Intensive 12-weeks full-time (40 hours per week) paid internship.

• Program start dates are in May.18 2026

• A stipend, based on location, will be provided to help alleviate costs associated with the internship. 

• Ownership of challenging and impactful business-critical projects.

• Work with some of the most talented people in the biotechnology industry.

Who You Are

Required Education

• Must be pursuing a PhD (enrolled student).

• Must have attained a PhD.