AI Developer | 2+ Years Experience | GenAI, LLM Fine-tuning, ComfyUI, Computer Vision
Actively Looking for Work
I’m an AI Developer
I’m an AI Developer with over two years of hands-on experience working at the intersection of computer vision, large language models (LLMs), and generative image systems. With a Master’s in Physics and a deep interest in applied AI, my work is focused on building intelligent design tools and automation systems for content-heavy domains like e-commerce. I specialize in transforming static product visuals into editable, dynamic templates using AI making the creative process faster, smarter, and more accessible.
Experience
At Zikhara AI
- I lead the development of a GenAI-powered design automation engine that integrates heuristics with reinforcement learning to make layout and styling decisions. This system ingests product assets, understands design context, and outputs ready-to-use A+ content.
- I’ve also built SVG parsers compatible with tools like Canva and Adobe, and led the development of a Photoshop plugin to allow seamless integration between design tools and AI pipelines.
On the LLM front
- I’ve fine-tuned open-source models like LLaMA and Mistral using Low-Rank Adaptation (LoRA) techniques, enabling them to perform domain-specific copywriting and layout logic tasks.
- I also integrate these models with frameworks like LangChain and LangGraph to build modular, scalable pipelines for information extraction, content generation, and semantic retrieval.
For generative image workflows
- I use ComfyUI extensively to solve inpainting, object placement, and consistency issues in product photography automation. This has allowed us to build realistic and reusable image assets at scale.
- I’ve also created intelligent scrapers with Playwright and LangChain that adaptively extract structured product data from ecommerce sites, a crucial input for automating both visuals and text.
Beyond my core projects
- I’ve contributed to NLP-focused tools like a multilingual text-to-speech app and an ethical, memory-aware chatbot for a women’s career foundation.
- The chatbot combined a semantic RAG pipeline with live job listings and incorporated gender bias detection, enabling inclusive and empowering interactions powered by Gemini and FAISS.
Passion
- I use ComfyUI extensively to solve inpainting, object placement, and consistency issues in product photography automation. This has allowed us to build realistic and reusable image assets at scale.
- I’ve also created intelligent scrapers with Playwright and LangChain that adaptively extract structured product data from ecommerce sites, a crucial input for automating both visuals and text.
Beyond my core projects
- I’ve contributed to NLP-focused tools like a multilingual text-to-speech app and an ethical, memory-aware chatbot for a women’s career foundation.
- The chatbot combined a semantic RAG pipeline with live job listings and incorporated gender bias detection, enabling inclusive and empowering interactions powered by Gemini and FAISS.
Passion
I’m deeply passionate about the intersection of AI and creativity especially how we can build systems that augment human designers and marketers rather than replace them. My long-term goal is to continue building AI tools that feel intuitive, context-aware, and genuinely useful for small teams, creators, and businesses.
Looking for Opportunities
I’m currently looking for a role as an AI Developer or AI Engineer where I can work on multimodal AI systems particularly projects involving generative image pipelines, LLMs, and creative automation. I'm especially excited about joining teams that are pushing the boundaries of what GenAI can do in real-world applications.