Machine Learning Infrastructure Engineer

About us slice A new bank for a new India slice’s purpose is to make the world better at using money and time, with a major focus on building the best consumer experience for your money. We’ve all felt how slow, confusing, and complicated banking can be. So, we’re reimagining it. We’re building every product from scratch to be fast, transparent, and feel good, because we believe that the best products transcend demographics, like how great music touches most of us. Our cornerstone products and services: slice savings account, slice UPI credit card, slice UPI, and slice business are designed to be simple, rewarding, and completely in your control. At slice, you’ll get to build things you’d use yourself and shape the future of banking in India. We tailor our working experience with the belief that the present moment is the only real thing in life. And we have harmony in the present the most when we feel happy and successful together. We’re backed by some of the world’s leading investors, including Tiger Global, Insight Partners, Advent International, Blume Ventures, and Gunosy Capital. About the team Everything great that AI touches, features extracted, models trained, pipelines running, LLMs responding becomes real when infrastructure makes it possible. We’re building the ML foundations layer that will power the next decade of intelligent banking. This means large-scale feature ETL, daily training pipelines, batch scoring jobs, online inference endpoints, feature stores and LLM infra designed so engineers and data scientists can deploy and iterate with reliability, speed and cost-efficiency. This team is the backbone of ML at slice. About the role We are looking for a founding Machine Learning Infrastructure Engineer to architect and scale our ML platform across offline, nearline, and online workloads. You will build and evolve the systems that power feature generation, model training, batch scoring and real-time inference for classical ML models and LLM-based applications. If you enjoy distributed systems more than models, love building platforms others depend on, and measure success in reliability, speed and developer productivity you’ll feel at home here. What you will do - Focus on scalability and cost effectiveness. - Design ML/LLM infrastructure across offline (batch), nearline (streaming) and online (API) paths - Own platform components for experimentation, deployment and model observability - Implement cost-optimised compute scaling, caching and spot fleet utilisation. - Build & maintain large-scale pipelines for feature ETL, batch training and scoring - Build systems for model packaging, versioning, lineage and rollback safety - Enable engineers to productionise model use cases at slice - Ensure data quality, validation, governance and secure access controls - Build abstractions so teams can deploy models faster with lower complexity What you will need - 5–10 years' experience in distributed systems / ML infra / pipeline engineering - Hands-on with Spark / Ray / Flink / Hadoop or similar compute engines - Understanding of cloud compute scaling (AWS/GCP/Azure) - Background in productionizing generative or open-weights models, tuning them for real-world performance, and ensuring reliable, predictable behaviour in production. - Experience with training or batch inference frameworks is a plus - Bonus: Kafka, feature stores, LLM infra, self-hosted LLMs and SLMs - Strong systems thinking, debugging ability and focus on reliability at scale - Ability to move fast while maintaining clarity, simplicity and stability Life at slice Life so good, you’d think we’re kidding: - Competitive salaries. Period. - Extensive health insurance for you & your dependents. - Flexible working hours, except 3 AM, we like to sleep. - Tailored leave policies so you never miss moments that matter. - Rewards that celebrate milestones, impact and ownership. - Learning & upskilling opportunities, seriously, not kidding. - Good food, games, and an office where colleagues feel like friends.