ML ops Engineer

Job Title:

Location:

Experience:

Industry:

About Us:

Why Join Us?

• Be at the forefront of AI innovation in finance technology with exposure to next-gen GenAI techniques.
• Collaborate with dynamic global teams spanning finance, technology, and client solutions.
• Work in a hybrid setup in Pune or Mumbai, blending flexibility with the spirit of teamwork.
• Opportunity to directly influence client-facing AI solutions impacting real-world business outcomes.
• Grow your career in an environment that champions best coding practices, continuous learning, and breakthrough AI deployments on cloud platforms.
  

What You’ll Do:

• Develop, and manage efficient MLOps pipelines tailored for Large Language Models, automating the deployment and lifecycle management of models in production.
• Deploy, scale, and monitor LLM inference services across cloud-native environments using - Kubernetes, Docker, and other container orchestration frameworks.
• Optimize LLM serving infrastructure for latency, throughput, and cost, including hardware acceleration setups with GPUs or TPUs.
• Build and maintain CI/CD pipelines specifically for ML workflows, enabling automated validation, and seamless rollouts of continuously updated language models.
• Implement comprehensive monitoring, logging, and alerting systems (e.g., Prometheus, Grafana, ELK stack) to track model performance, resource utilization, and system health.
• Collaborate cross-functionally with ML research and data science teams to operationalize fine-tuned models, prompt engineering experiments, and multi agentic LLM workflows.
• Handle integration of LLMs with APIs and downstream applications, ensuring reliability, security, and compliance with data governance standards.
• Evaluate, select, and incorporate the latest model-serving frameworks and tooling (e.g., Hugging Face Inference API, NVIDIA Triton Inference Server).
• Troubleshoot complex operational issues impacting model availability and degradation, implementing fixes and preventive measures.
• Stay up to date with emerging trends in LLM deployment, optimization techniques such as quantization and distillation, and evolving MLOps best practices.
  

What We’re Looking For:

• 3 to 5 years of professional experience in Machine Learning Operations or ML Infrastructure engineering, including experience deploying and managing large-scale ML models.
• Proven expertise in containerization and orchestration technologies such as Docker and Kubernetes, with a track record of deploying ML/LLM models in production.
• Strong proficiency in programming with Python and scripting languages such as Bash for workflow automation.
• Hands-on experience with cloud platforms (AWS, Google Cloud Platform, Azure), including compute resources (EC2, GKE, Kubernetes Engine), storage, and ML services.
• Solid understanding of serving models using frameworks like Hugging Face Transformers or OpenAI APIs.
• Experience building and maintaining CI/CD pipelines tuned to ML lifecycle workflows (evaluation, deployment).
• Familiarity with performance optimization techniques such as batching, quantization, and mixed-precision inference specifically for large-scale transformer models.
• Expertise in monitoring and logging technologies (Prometheus, Grafana, ELK Stack, Fluentd) to ensure production-grade observability.
• Knowledge of GPU/TPU infrastructure setup, scheduling, and cost-optimization strategies.
  

• Effective communication and collaboration skills to work with cross-functional teams in a fast-paced environment.
  

Educational Background:

• Bachelor’s or Master’s degree from premier Indian institutes (IITs, IISc, NITs, BITS, IIITs etc.) in:
• Computer Science, or
• Any Engineering discipline, or
• Mathematics or related quantitative fields.
  

Benefits:

• Hybrid work model combining remote flexibility and collaborative office culture in Pune or Mumbai.
• Continuous learning budget for certifications, workshops, and conferences.
• Opportunities to work on industry-leading AI research and shaping the future of financial services.