Skip to main content

Building with impact: 2025 Dropbox intern engineering projects

6Jdcloti
This summer, the Emerging Talent team proudly welcomed 43 interns to Dropbox as part of our 2025 Camp Dropbox Intern Program. Representing 27 colleges and universities—including six international institutions in Canada, Poland, and Ireland—this year’s cohort brought a wealth of diverse perspectives and experiences. Of the group, 28 interns joined our Engineering teams, and over the course of 12 weeks (May through September), they immersed themselves in meaningful work, continuous learning, and our Virtual First culture.
The Dropbox Intern Program is thoughtfully designed to cultivate growth, spark innovation, and build lasting connections. Interns benefited from more than 6,000 hours of dedicated one-on-one mentorship, tackled high-impact projects aligned with team and company goals, and explored hands-on applications of AI through tools like CoPilot and ChatGPT. Many of these projects supported the development of Dropbox Dash, our AI-powered universal search product, while new initiatives such as the Flow Club pilot supported deep focus. Robust programming—including Virtual First events, ERG activities, and the in-person Emerging Talent Summit—created further opportunities for connection and community. By the end of the summer, these interns had made meaningful contributions across our engineering organization.
Below, our interns share what they worked on this summer, from big technical wins to moments of creativity, collaboration, and growth that shaped their time at Dropbox.
“I built the GrowthBook_on_Web workflow, a new evaluation layer that lets Dropbox.com run GrowthBook and our legacy Stormcrow flags side by side. The hybrid path picks the fastest engine per feature, trims p95 flag-evaluation latency from 12 ms to under 3 ms, and removes a 70 MB server call on every page. When fully rolled out, it’ll accelerate load times for hundreds of millions of visitors and give engineers, PMs, and designers a single cross-product experimentation switchboard spanning Dropbox, HelloSign, and DocSend. Faster feedback loops mean ideas ship sooner—and customers feel the improvement with every click.”
—Allan Zhou, Platform Experimentation
“I tackled the Dropbox file history tracking system, which cost $2M annually due to redundant FileIDEntity logs in Edgestore. By redesigning the workflow to leverage our modern, tree-based Cypress data model with stable Object IDs, I built a new RPC API that queries file revision history directly from our shard database. This eliminates data redundancy while maintaining functional parity with the legacy system from the user's perspective. As an engineering intern working in a large production database for the first time, I learned the importance of strongly tested, verifiable code and thoughtful system design. I really aligned with the core Dropbox values of Be Worthy of Trust and Keep It Simple at the software level. This solution simplifies our metadata infrastructure, significantly reduces operational costs, and shows how thoughtful refactoring of legacy systems can deliver both technical elegance and substantial business value.”
—Rhea Rai, Filesystem Data
“During my internship with the ML Platform team, I developed AI Sentinel, a system that monitors the health of ML model deployments and synchronizes this data to Weights & Biases, our new model training platform. By integrating with internal inference services, AI Sentinel gives machine learning engineers (MLEs) real-time operational visibility they previously had to gather manually. The result is greater deployment confidence and faster iteration cycles, ensuring reliable ML model deployments that power Dash’s intelligent features at scale.”
—Ben Juntilla, ML Platform
“I worked on reducing front end latency in Magic Pocket. Elevated PUT latencies during scheduled disk restarts can delay updates in workflows like Dash connectors, leaving users with outdated or missing content. To address this, I built a cache to track storage health and added a filtering option to skip degraded volumes. This health-aware routing reduces slow writes and gives operators greater control, ensuring Dropbox delivers timely, accurate search results. By the end of the project, I cut promotion latency by 20% in our staging zones and identified blockers that, once resolved, will unlock further improvements.”
—Albert Joon Sung, Storage Core
“This summer, I worked on adding search suggestions to the Dropbox desktop app. The project aligned closely with core Dropbox values of usability and delivering a smooth user experience, as search suggestions help users find their files more quickly and easily. To ensure a modular, readable, and reusable implementation, I created an implementation document before jumping into coding. At each step, I consulted with my team to make sure my code was clean and efficient, and I documented my work thoroughly to guide future developers who build on this project. As the feature rolls out, I hope it enhances the search experience for all Dropbox desktop users and encourages greater engagement with the search feature.”
—Elena Scobici, Desktop Core Experience
“Dropbox sharing is technically complex, but my intern project focused on simplifying the user experience. For customers who value confidentiality, it’s crucial to know who can access their documents. The improved copy link experience makes this easier by allowing users to update and automatically save certain link settings—streamlining the process of creating the perfect link. Working on this feature reinforced Dropbox values of They Win, We Win and Make Work Human. I designed the project with a modular, composable architecture that streamlined development and made small changes simple to implement. AI tools also accelerated my onboarding, helping me navigate the vast codebase and grow quickly as an independent engineer.”
—Michelle Sotelo, Sharing Experience
“I worked on our dynamic application security testing (DAST) system as part of the Application Security team. DAST scans web endpoints for vulnerabilities and simulates real attackers to verify that our systems can withstand cyber threats. My work involved adapting open-source scanning tools to the Dropbox production environment, configuring them to detect a wide range of web security issues, and defining best practices so all Dropbox developers can build with security in mind. The DAST system is now live for Dropbox Core, with plans to expand to Dash and other Dropbox products in the near future.”
—Shreyas Mayya, Application Security
“S3AL_Console is an internal Dropbox tool that simplifies S3AL storage management and onboarding. It offers a streamlined UI for navigating buckets and folders, along with a few-click flow for setting up new S3AL buckets. Designed to reduce support overhead and encourage adoption, the console improves efficiency and unlocks access to previously unusable data, such as multipart objects and cross-store reconciliation of bucket contents. By giving users clear tools to understand their buckets and create new ones, S3AL_Console makes it easier and more appealing for Dropbox teams to migrate to S3AL and leverage Object Store as a lower-cost alternative to S3.”
—William Bolduc, Storage Platform 
“I worked on Autobatross, an AI-powered tool built on top of our internal Albatross migration platform. Autobatross automates code migrations by letting developers launch Autocode jobs on specific folders for specific migration types. If a migration succeeds, it automatically generates a PR; if not, developers can run the command locally and submit changes manually. The tool is fully customizable, whether used through the CLI or as part of an automated workflow. Using Autobatross, I was able to complete two major migrations.”
—Ahmed Ibrahim, Web Developer Experience
“I built tools that give machine learning engineers access to the most up-to-date information in the Dash persistence store. With this, downstream teams can train models on fresher data and pull in additional metadata fields from third-party systems without waiting for the Connector Platform Team to redownload or repackage anything.”
—Eddie Ormseth, Connector Platform
“I worked on expanding the unified search platform (USP) to support more than 20 languages. USP powers search across Dropbox products like Replay, and my project integrated a language detection pipeline into both indexing and retrieval. This enables accurate, efficient multilingual search without the overhead of traditional solutions. By delivering native language support ahead of Dash’s upcoming launch, my work helps Dropbox scale globally, improve the developer experience, and unlock richer search for international customers—bringing us closer to our vision of an AI-first, universally accessible search platform.”
—Rishi Peddakama, Retrieval Platform
“I explored advanced anomaly detection techniques for our Vortex2 metrics system. Traditional static alerting can miss sudden, meaningful shifts in data that don’t cross predefined thresholds (or trigger too often when changes are expected). To address this, I developed adaptive detection methods that adjust to evolving patterns. These improvements streamline alert creation and reduce alert fatigue, enhancing the on-call experience. By accounting for seasonality, the new anomaly detection functions also enable faster response times and improve the overall developer experience.”
—Yonatan Ginsburg, Metrics
“I developed a seamless document preview experience within Dropbox Dash, allowing users to quickly view file content without leaving their search context. This enhancement supports the Dropbox mission to accelerate workflows by reducing context switching and increasing engagement. I built interactive UI components, integrated PDF viewing, and implemented dynamic follow-up features linking to AI-powered chat. Working closely with partner teams, I ensured secure file access and added telemetry tracking to support a data-driven rollout. The project improves the actionability of search results, helping users verify documents faster and engage more deeply with Dash features.”
—Francesca Venditti, Find & Discover
“My internship focused on prototyping a machine learning–based approach to dynamically tune H.264 encoding parameters for Dropbox video previews. I collected and labeled video data, engineered features, and trained regression models to predict efficient encoder settings based on content characteristics. After integrating the model into a staging environment, I ran A/B-style benchmarks to measure impacts on quality and processing time. The results showed promising reductions in encoding cost with minimal perceptible quality loss. While not yet deployed, the project provides a roadmap for future adoption and offers a path to scaling preview generation more sustainably.”
—Pablo Velarde, Content Processing
“This summer on the Analytics Platform team, I worked on optimizing large-scale Databricks queries and ETL pipelines to reduce compute cost and latency. I developed an optimization recommendation system that flagged high-cost query patterns, expensive table-column filters, and under-allocated compute resources, complete with actionable sourcing information. Using these insights, I tested liquid clustering on multi-terabyte datasets, achieving up to 50% runtime reductions for targeted queries and projecting annual savings of over $100K. I also prototyped and documented an Airflow pipeline to migrate a 500TB mobile events log to liquid clustering, paving the way for broader adoption of modern data layout techniques.”
—Sanjith Udupa, Analytics Platform
“I built an automated agent, a system that automatically detects, uploads, and triages MySQL core dumps across our database fleet. MySQL crashes can expose durability risks and potential user data loss, so early detection is critical to maintaining our integrity guarantees. The agent I worked on monitors for new crashes, extracts stack traces and metadata to Sentry for AI-driven triage, and leverages Sentry’s hashing algorithm to classify unique crashes. Full dumps are persisted to S3AL for long-term analysis. By improving observability into database failures, coredumpagent helps us identify recurring issues earlier, prevent future SEVs, and strengthen our core infrastructure reliability.”
—Abdul Cherchar, Metadata Core
“I integrated Dropbox with Google’s Cross-Account Protection (CAP) using the RISC standard to process real-time alerts of compromised Google accounts. I built a production-ready ingestion pipeline that validates and parses CAP event tokens with structured logging, caching for low latency, and deduplication to handle retries. A publisher/subscriber async system processes events concurrently and triggers automated defenses such as session invalidation, app logout, and MFA enforcement. Since over 60% of Dropbox logins occur via Google, the system—now deployed to all Google sign-in users—advances our 2025 priority to strengthen account security for both admins and users.”
—Anna Yang, account Experience
“I developed a feature that lets Dropbox admins exclude known safe files from malware alerts. False positives can overwhelm admins, leading to alert fatigue and reduced trust in security tools. To solve this, I built a scalable backend system with a database for storing exclusions, updated the frontend with intuitive controls, and partnered with designers to refine the experience. I also added logging, metrics, and thorough test coverage to ensure reliability. This work gives admins greater control, reduces noise in our security systems, and strengthens our ability to deliver trustworthy protection at scale.”
—Bartosz Brejna, Admin Security
“I built an extensible AI web-automation agent for Dropbox. It lets developers create plug-in ‘skills’ that automate browser tasks, powered by a ReAct loop that plans, interacts with the page, and learns from what it sees. I also connected Dropbox backend APIs via searchFile and uploadFile actions to fetch and upload files, using open-source foundations. By keeping tool sets small and modular, developers can quickly compose reliable, task-specific automations—like form filling or proofreading. As demand for automating repetitive web tasks continues to grow, integrating automation tools into Dash will significantly improve the user experience.”
—Alan Zhu, Conversational AI
~ ~ ~ 
If building innovative products, experiences, and infrastructure excites you, come build the future with us! Visit jobs.dropbox.com to see our open roles.

Your team, your impact

We've revolutionized cloud storage. Now, we're pioneering the future of AI-enhanced work. See how your potential can unlock something bigger at Dropbox.

For a personalized experience, let’s start with your

ZIP Code

Enter your ZIP code