Narayanan Sivakumar
Ai Ml Engineer @FDM Group
Signup · Get unlimited contacts
WORK HISTORY
Ai Ml Engineer @FDM Group
Built AI agents from the ground up using LangChain and LangGraph, designing multi-step workflows that combined perception, reasoning, and action execution, reducing manual task handling by 35% in internal test environments.• Implemented stateful, graph-based agent workflows to manage memory, retries, and decision paths, improving agent reliability and reducing failure loops by 30% during complex multi-step tasks.• Developed and trained machine learning models using PyTorch, TensorFlow, and Scikit-learn, enabling agents to learn from historical data and adapt their behavior, improving task success rates by 20%.• Integrated AI agents with external APIs, databases, and simulation environments, allowing agents to fetch data, trigger actions, and store memory autonomously, cutting human intervention time by 40%.• Designed reinforcement learning and planning logic for agent decision-making, helping agents choose optimal actions in uncertain or dynamic environments and improving long-horizon task completion by 25%.• Collaborated closely with data scientists and software engineers to test and refine agent behaviors, using feedback from simulations and user testing to improve performance and stability across releases.• Ran controlled experiments and simulations to evaluate agent performance across accuracy, latency, and robustness metrics, helping identify edge cases and improve production readiness by 30%.• Optimized agent workflows for performance and scalability, reducing inference latency by 20% and ensuring agents remained responsive under higher workloads.• Containerized agent services using Docker and supported deployments to cloud platforms (AWS / Azure / GCP), ensuring consistent behavior across development and production environments.• Documented agent architectures, design decisions, and workflows clearly, enabling reproducibility and reducing onboarding time for new team members by 40%.
EDUCATION
SRM IST Chennai
Bachelor of Technology - BTech, Computer Science
University of Alabama at Birmingham
Master's degree, Computer Science
ABOUT NARAYANAN SIVAKUMAR
AI/ML Engineer with 4+ years of experience designing, developing, and deploying production-ready machine learning systems, AI agents, and scalable ML platforms across enterprise and cloud environments.• Demonstrated expertise in building agentic AI workflows using LangChain and LangGraph, delivering reliable multi-step reasoning systems that reduced manual effort by 35–40% and improved workflow stability by 30%.• Strong end-to-end ownership of the ML lifecycle, including data preparation, model training, evaluation, deployment, and monitoring, resulting in 30–40% faster release cycles and consistent performance improvements of 20–25% across models.• Extensive hands-on experience with PyTorch, TensorFlow, Scikit-learn, Spark, and distributed ML pipelines, enabling scalable experimentation, efficient training, and high-throughput inference in production environments.• Proven ability to design and operate ML infrastructure and MLOps pipelines using Airflow, MLflow, Kubeflow, Docker, and Kubernetes, automating workflows and reducing manual operational overhead by 35%+.• Experienced in building RESTful ML services and microservices in Python and Java, accelerating integration of AI capabilities into business systems and improving adoption across teams.• Collaborative engineer who works closely with ML scientists, data engineers, product managers, and platform teams to translate research and experimentation into robust, production-grade AI solutions.• Deep interest in AI agents, reinforcement learning, and scalable system design, with a consistent focus on reliability, performance optimization, and long-term maintainability of ML solutions in production.
This profile is compiled from publicly available professional sources. Unifers is not affiliated with or endorsed by LinkedIn. Request removal of this profile.