Alexander Frömmgen

Gheorghe Comanici, Eric Bieber, Mike Schaekermann et al. · amazon-science, baidu

In this report, we introduce the Gemini 2.X model family: Gemini 2.5 Pro and Gemini 2.5 Flash, as well as our earlier Gemini 2.0 Flash and Flash-Lite models. Gemini 2.5 Pro is our most capable model yet, achieving SoTA performance on frontier coding and reasoning benchmarks. In addition to its incredible coding and reasoning skills, Gemini 2.5 Pro is a thinking model that excels at multimodal understanding and it is now able to process up to 3 hours of video content. Its unique combination of long context, multimodal and reasoning capabilities can be combined to unlock new agentic workflows. Gemini 2.5 Flash provides excellent reasoning abilities at a fraction of the compute and latency requirements and Gemini 2.0 Flash and Flash-Lite provide high performance at low latency and cost. Taken together, the Gemini 2.X model generation spans the full Pareto frontier of model capability vs cost, allowing users to explore the boundaries of what is possible with complex agentic problem solving.

Daye Nam, Ahmed Omran, Ambar Murillo et al.

Large Language Models (LLMs) are rapidly transforming software engineering, with coding assistants embedded in an IDE becoming increasingly prevalent. While research has focused on improving the tools and understanding developer perceptions, a critical gap exists in understanding how developers actually use these tools in their daily workflows, and, crucially, where they struggle. This paper addresses part of this gap through a multi-phased investigation of developer interactions with an LLM-powered code editing and transformation feature, Transform Code, in an IDE widely used at Google. First, we analyze telemetry logs of the feature usage, revealing that frequent re-prompting can be an indicator of developer struggles with using Transform Code. Second, we conduct a qualitative analysis of unsatisfactory requests, identifying five key categories of information often missing from developer prompts. Finally, based on these findings, we propose and evaluate a tool, AutoPrompter, for automatically improving prompts by inferring missing information from the surrounding code context, leading to a 27% improvement in edit correctness on our test set.

Vincent Nguyen, Guilherme Herzog, José Cambronero et al.

Manually editing pasted code is a long-standing developer pain point. In internal software development at Google, we observe that code is pasted 4 times more often than it is manually typed. These paste actions frequently require follow-up edits, ranging from simple reformatting and renaming to more complex style adjustments and cross-language translations. Prior work has shown deep learning can be used to predict these edits. In this work, we show how to iteratively develop and scale Smart Paste, an IDE feature for post-paste edit suggestions, to Google's development environment. This experience can serve as a guide for AI practitioners on a holistic approach to feature development, covering user experience, system integration, and model capabilities. Since deployment, Smart Paste has had overwhelmingly positive feedback with a 45% acceptance rate. At Google's enterprise scale, these accepted suggestions account substantially for over 1% of all code written company-wide.

Jonas Höchst, Artur Sterz, Alexander Frömmgen et al.

We present a novel data-driven approach to perform smooth Wi-Fi/cellular handovers on smartphones. Our approach relies on data provided by multiple smartphone sensors (e.g., Wi-Fi RSSI, acceleration, compass, step counter, air pressure) to predict Wi-Fi connection loss and uses Multipath TCP to dynamically switch between different connectivity modes. We train a random forest classifier and an artificial neural network on real-world sensor data collected by five smartphone users over a period of three months. The trained models are executed on smartphones to reliably predict Wi-Fi connection loss 15 seconds ahead of time, with a precision of up to 0.97 and a recall of up to 0.98. Furthermore, we present results for four DASH video streaming experiments that run on a Nexus 5 smartphone using available Wi-Fi/cellular networks. The neural network predictions for Wi-Fi connection loss are used to establish MPTCP subflows on the cellular link. The experiments show that our approach provides seamless wireless connectivity, improves quality of experience of DASH video streaming, and requires less cellular data compared to handover approaches without Wi-Fi connection loss predictions.