Software Engineer, GeminiApp Info Seek Quality at Deepmind

Snapshot We are the Information Verticals team for the Gemini App, bringing search, maps, shopping and other information-related features to both the web and mobile Gemini applications. We are a fast-paced organization bringing features from ideation to launch in a matter of months. We are a non-bureaucratic, innovative, bottom-up organization. We are seeking an experienced and self-motivated Software Engineer with particular expertise in quality and AI applications to ensure the Gemini applications deliver world-class, reliable, and personalized experiences. About Us Within the Information Verticals team, the Shopping team is at the forefront of Google's shift towards agentic experiences. We are redefining how users discover, research, and purchase products by combining the reasoning capabilities of Gemini with the comprehensive data of the Shopping Graph. To constantly innovate, we work closely with various Google DeepMind (GDM) research units. Following a breakout year with two major launches in Q4 (launch announcement) that were well received in the press, we have established a new baseline for highly visual agentic shopping. We have one of the highest favorability & retention metrics on GeminiApp have and were featured prominently in the recent "Personal Intelligence" launch (launch, Sundar tweet). This role offers the unique opportunity to make impact on a high-growth vertical that is already seeing massive traction,). The Role Join Google DeepMind's mission to advance science and benefit humanity by solving intelligence. We are seeking an experienced and self-motivated Software Engineer with a passion for quality to ensure the Gemini applications deliver world-class, reliable, and personalized experiences. In this role, you will be building new features in the Gemini product, focused on shopping and other information-related verticals like search, maps & news. You’ll prototype new LLM applications, design and implement evaluation metrics, collection evaluation data, analyze model responses, and conduct prompt engineering in order to ship high-quality features to our users. This is a rare opportunity to shape a new era of technology, accelerating development with the latest AI models and collaborating with AI to build high-fidelity user experiences. You will see your work in the hands of users frequently and quickly. As a key contributor to one of Google's top efforts for 2025, you will help shape the future of AI. What You'll Do • Implement sophisticated LLM-powered systems, including context engineering, agentic workflows with tool use, and robust evaluation frameworks. • Dive deep into model behavior, creating high-quality, diverse, and challenging evaluation datasets to identify weaknesses and guide improvements. • Make improvements to the serving and evaluation infrastructure to power new features. • Develop and scale evaluation solutions (auto-raters, human evals) to continuously monitor and improve the performance of LLM-powered features. • Drive collaboration with cross-functional partners (Product, UX, Research) to translate user needs into intuitive and valuable features within the Gemini ecosystem. • Collaborate with GenAI research teams to integrate product-driven data and evaluations to the base Gemini models. • Analyze user data and model outputs to enhance the quality of personalized experiences, ensuring Gemini is helpful and relevant for every user. • Champion a culture of quality, defining metrics, setting standards, and driving best practices for LLM evaluation and responsible AI development. About You We seek out individuals who have the skill and passion to design and build high-quality components, while simultaneously being willing to move fast and pivot when necessary. We are looking for strong and adaptable engineers who are obsessed with quality and have a deep understanding of how to measure and improve AI systems. You are a flexible problem-solver, comfortable inventing novel solutions and quickly pivoting when ideas need to change. In order to set you up for success as a Software Engineer on GeminiApp, we look for the following skills and experience: • User-centric vision and passion for shipping best products to our users, with a strong track-record of shipping multiple successful consumer features. • Experience working with a small team in a startup-like environment with rapidly changing priorities as product needs evolve • A deep understanding of Generative AI and Large Language Models. • Direct experience with building evaluation datasets, LLM evaluation techniques, human rating paradigms, or building automated raters. • Proven experience in loss analysis, quality hillclimbing, prompt engineering, and optimization. • Strong background in data analysis, experimental design, and statistical methods. • Proficiency in Python, having experience with C++ is a plus. • Excellent communication and collaboration skills, with the ability to translate complex quality issues into actionable plans for diverse, cross-functional teams. In addition, the following would be an advantage: • Experience with common ML frameworks (e.g., PyTorch, JAX, TensorFlow). • Familiarity with user metrics logging and A/B testing for ML-powered products. • Proven knowledge and experience building scalable, efficient C++ backend infrastructure components within a complex distributed system. • Experience with standard Google technologies including Boq, Stubby, StreamZ, Mendel and Sawmill. The US base salary range for this full-time position is between $197,000 - $291,000 + bonus + equity + benefits. Your recruiter can share more about the specific salary range for your targeted location during the hiring process. Note: In the event your application is successful and an offer of employment is made to you, any offer of employment will be conditional on the results of a background check, performed by a third party acting on our behalf. For more information on how we handle your data, please see our Applicant and Candidate Privacy Policy. At Google DeepMind, we value diversity of experience, knowledge, backgrounds and perspectives and harness these qualities to create extraordinary impact. We are committed to equal employment opportunities regardless of sex, race, religion or belief, ethnic or national origin, disability, age, citizenship, marital, domestic or civil partnership status, sexual orientation, gender identity, pregnancy, or related condition (including breastfeeding) or any other basis as protected by applicable law. If you have a disability or additional need that requires accommodation, please do not hesitate to let us know. For fairness reasons, Recruiters and Hiring Managers do not offer informational calls. If you are interested in learning about the role, please share a copy of your CV and a cover letter with the Recruiter listed via email (not G-Chat) and they will process your application. Please avoid reaching out to the Hiring Manager directly, as this can cause delays in processing your application. You will receive updates on the status of your application through our standard recruitment channels, and if you are shortlisted for the position, we will contact you to set up an initial interview where you will have the opportunity to ask questions and learn more about the position.