I started in applied ML for energy and IoT — forecasting consumption, wiring devices into analytics platforms, and learning that models only matter when data pipelines and operations hold up in production. That work at ZeroTEnergy taught me MLOps discipline and how physical systems generate the signal ML needs.
At CarbonCompete I stepped into product leadership for climate/ESG SaaS: schema design, collection workflows, and LLM features that made messy enterprise data usable. We improved collection by ~90% and patented LLM-based handling for advanced analysis — a clear lesson that GenAI is highest leverage when it sits inside a real product loop.
Since then I’ve doubled down on tool-using LLMs and multi-cloud delivery: knowledge assistants and agents at DoubleKlick (Bedrock, Vertex AI, DigitalOcean AI), open-source MCP servers that connect Claude-class hosts to LinkedIn, Medium, Kaggle, and X, and now Senior Data Scientist work at NuSummit in Gurugram combining GenAI with enterprise and sustainability outcomes. I’m looking for senior Applied GenAI roles onsite in Gurugram — grounding, tool contracts, evaluation, and ships that survive beyond demos.
