Ganesh Khovalan

AI Systems Engineer specializing in MLOps and Open Source LLM orchestration. I bridge the gap between theoretical Machine Learning and production-grade software. My core expertise lies in reducing inference latency and cost by deploying quantized models (Llama 3, Mistral) via Ollama and vLLM. With a strong foundation in Mechatronics logic and recent professional experience in Generative AI automation, I engineer "Agentic" systems that are robust, scalable, and privacy-focused.

AI Engineer at RootedAI (2025-02 – 2025-10)

• WhatsApp Automation (Pet Food Sector): Architected a high-throughput conversational agent that handles the full customer lifecycle from query to payment.
• ML Engineering: Curated a custom dataset to fine-tune a Llama-3-8B model using QLoRA, improving domain understanding for pet nutrition by 35%.
• Transactional Architecture: Integrated Razorpay API webhooks for real-time payment verification, enabling the agent to autonomously finalize e-commerce orders and close sales within the chat interface.
• AI Photography Pipeline: Built a localized GenAI workflow for automated product asset generation.
• Computer Vision: Implemented a Stable Diffusion pipeline with custom-trained LoRA adapters to generate brand-consistent product images, enforcing strict style guidelines without manual prompting.

Testing & Validation Engineer at Konwert India Motors Pvt Ltd (2024-04 – 2024-08)

• Data Engineering: Engineered Python pipelines to decode high-frequency CAN bus telemetry, transforming raw binary streams into structured datasets for anomaly detection.
• System Logic: Applied embedded control logic principles to optimize data throughput—experience I now leverage to optimize asynchronous AI inference pipelines.

B.E. in Mechatronics Engineering – Sri Krishna College of Engineering and Technology (2021 – 2025)