FAQ

Common questions about my work

I'm a Full Stack Developer with 8+ years of experience specializing in React.js, Node.js, Python, and AI/ML technologies. I've worked with companies like Consult Anubhav and WeEnggs Technology, building everything from AI-powered assistants to enterprise web applications.

I'm based in Bharuch, Gujarat, India. I work both remotely and on-site depending on project requirements and have experience collaborating with teams across different time zones.

My core expertise includes Frontend (React.js, Redux.js, HTML5, CSS3, TypeScript, TailwindCSS, SwiftUI), Backend (Node.js, Python, PHP, Go, TypeScript), Databases (MongoDB, MySQL, Firebase), AI/ML (Generative AI, Agentic AI, OpenAI, Groq, Ollama), and DevOps (AWS, Azure, Docker, CI/CD, GitHub Actions).

Absolutely! I specialize in AI/ML technologies including Generative AI, Agentic AI with tool calling, offline AI solutions, and have built projects like Ubik AI (personal AI assistant) and AI Transcribe (meeting transcription with multi-language support).

I work on diverse projects including on-premise AI deployment for privacy-sensitive clients, custom MCP server development, LLM finetuning pipelines, SaaS product development and team leadership, Google Workspace add-ons, client-specific Chrome extension automation, AI-powered applications and chatbots, E-commerce platforms and web applications, Chrome extensions and browser automation, and Data visualization and analytics tools.

I work on both new projects from scratch and existing codebases. I'm experienced in code refactoring, performance optimization, feature additions, and system migrations.

My AI work goes deeper than API integration. I build on-premise LLM pipelines for clients who can't use cloud APIs — including finetuning (QLoRA/LoRA), GGUF export, and local Ollama deployment. I build custom MCP servers that let LLM agents call any internal tool or API. I design agentic workflows that handle multi-step automation where off-the-shelf tools like n8n fail. I also build and deploy my own finetuned models (see YubiPic). If you need AI that actually works inside your infrastructure — not a wrapper around OpenAI — that's my speciality.

I follow agile methodologies with initial consultation and requirement analysis, technical architecture planning, iterative development with regular updates, testing and quality assurance, deployment and post-launch support, and documentation and knowledge transfer.

I maintain regular communication through your preferred channels (email, WhatsApp, video calls). I provide regular progress updates, share development milestones, and ensure transparency throughout the project.

Timelines vary based on project complexity: Simple websites/apps (1-2 weeks), Medium complexity projects (3-6 weeks), Complex AI/enterprise solutions (2-3 months). I always provide detailed estimates after understanding your requirements.

Yes — I've mentored 20+ developers over my career and am currently leading a team of 6–10 junior developers across two production SaaS platforms (WhatsEase and Krosspost.ai). I also create educational content on YouTube and write technical articles on dev.to.

You can connect with me through various channels: **Social Media:** - LinkedIn: https://linkedin.com/in/yash-d-desai - Twitter (X): https://x.com/yash_d_desai - Instagram: https://www.instagram.com/yashdes.ai/ - Dev.to: https://dev.to/yashddesai/ - GitHub: https://github.com/yash-desai-naik - Linktree: https://linktr.ee/yashddesai **Email:** contact@yashddesai.com **WhatsApp:** Check my contacts page for WhatsApp details **Location:** I'm based in Bharuch, Gujarat and also serve Vadodara, Surat, and nearby cities. I work remotely worldwide.

My Linktree at https://linktr.ee/yashddesai is your one-stop hub for all my professional links and social media profiles. It includes direct links to my portfolio, GitHub repositories, LinkedIn profile, dev.to articles, Instagram, Twitter, and other platforms. This makes it easy to find and connect with me on your preferred platform without having to search individually.