Headline

Senior Data Scientist | ML & LLM Systems | Time-Series, Anomaly Detection, Agentic AI | Gold Medalist (MCA, VIT) | Building production-grade AI systems

About

👋 Hi, I'm Akash Kumar, a Senior Data Scientist with 3+ years of experience building and deploying production machine learning systems in the power and solar energy domain. I specialize in developing real-time ML solutions, including anomaly detection, predictive analytics, and automated data pipelines for industrial systems. I hold a Gold Medal in MCA from Vellore Institute of Technology and have hands-on experience working with time-series data, transformer models, and LLM-based applications. Core Stack Python • SQL • Machine Learning • Time-Series Analytics Flask • REST APIs • Docker • Kubernetes • MQTT LLMs • RAG • LangChain • Transformers I focus on building practical, scalable AI systems that convert complex data into actionable insights.

Location

Bengaluru

AI stack

antigravity

codex

Work experience

Senior Data Scientist at ExactSpace, Apr 2026 - Present · 5 mos, Hybrid

Data Scientist at ExactSpace, Jun 2023 - Mar 2026 · 2 yrs 10 mos, On-site, • Architected and deployed a zero-config, self-service ml orchestration platform enabling non-technical users to train, benchmark 10+ algorithms, and deploy models via ui, slashing development cycles by 80%. • engineered transformer-based autoencoders for real-time anomaly detection on high-frequency time-series operational data, enabling proactive industrial system maintenance. • built scalable flask-based ml apis for real-time inference, reducing latency and enabling sub-second anomaly detection for mission-critical assets. • implemented robust real-time data ingestion pipelines using mqtt for continuous monitoring and analytics. • designed automated ml pipelines featuring drift detection, auto-retraining, and performance monitoring to ensure model reliability. • containerized ml services with docker and managed deployments on kubernetes clusters for high availability. • developed automated reporting systems using python to generate client-ready pdf/excel insights. • mentored two adani power engineering teams (12 members) in building ml-based solutions, identifying potential cost savings of 2.5 crore/year.

Data Science Intern at ExactSpace, Aug 2022 - May 2023 · 10 mos, On-site, • Supported development of ML-based anomaly detection systems for industrial monitoring. • Built and debugged REST APIs connecting ML models with production systems. • Worked with MQTT-based microservices for real-time alerts and monitoring systems.

Education

Master of Computer Applications - MCA, Computer Science, Vellore Institute of Technology

BCA, Computer Science, Makhanlal Chaturvedi National University of Journalism and Communication, Bhopal