Rushil Dhube

AI Researcher

Ethosh Digital • Full time

Nov 2025 - Present • 6 mos

I work as an AI Automation & Backend Engineer at Ethosh Digital, Baner, where I design and implement intelligent, production-grade systems that integrate artificial intelligence into real-world workflows. My role combines AI engineering, workflow automation, and backend system development, with a focus on building scalable, reliable, and maintainable AI-driven solutions. A key part of my work involves creating end-to-end automation pipelines using n8n, orchestrating complex workflows that connect APIs, databases, third-party platforms, and AI models. These automations support event-driven processes, scheduled jobs, data processing, AI inference, and downstream actions, enabling faster execution and reduced manual effort. I also specialize in building AI wrappers and abstraction layers that standardize how AI models are used across applications. These wrappers handle prompt management, inference routing, response validation, error handling, logging, and monitoring, making AI capabilities reusable and easier to scale across multiple systems. In addition, I work with deep learning models, integrating pretrained and fine-tuned models for tasks such as text analysis, document intelligence, and image processing. I focus on deploying these models as reliable services rather than isolated experiments. On the backend, I develop core APIs and services that support AI workloads, including model-serving endpoints, authentication, background jobs, and performance optimization. Overall, my work centers on turning AI into practical, automated, and scalable solutions that deliver measurable business impact.

AI Intern

SniperThink • Full time

Aug 2025 - Nov 2025 • 3 mos

Independently led and managed the end-to-end social media automation function from inception to execution, while contributing to the development, optimization, and maintenance of advanced voice and chat agent workflows. Successfully architected and deployed comprehensive automation systems encompassing content generation, strategic multi-platform publishing, complex API integrations, and creative asset production by leveraging state-of-the-art generative AI tools and modern automation frameworks. Played a pivotal role in supporting conversational agent pipelines across speech-to-text, text-to-speech, speech-to-speech, and telephony integrations. Gained extensive hands-on expertise in growth systems architecture, automation infrastructure design, and startup scaling through active involvement in technical architecture planning, strategic product development, MVP cycles, and rapid iterative deployment. Key Contributions: - Assumed full ownership of the social media automation stack, managing all phases from strategy and workflow design to building, deploying, and continuously monitoring intelligent content generation systems with seamless multi-platform publishing across diverse social channels - Designed and implemented sophisticated automation workflows using Make.com to orchestrate complex multi-step processes, while developing custom direct integrations with the Meta Graph API using production-grade code to surpass platform limitations and enable advanced functionality beyond standard interfaces - Automated creative asset production by strategically utilizing cutting-edge generative AI tools, including Google Gemini for high-quality image creation and Google Veo 3.1 for professional-grade video production, significantly streamlining content creation pipelines for social media campaigns - Maintained and optimized robust voice and chat agent pipelines incorporating advanced technologies to ensure high performance, scalability, and reliability