About

About Arpit Desai

I’m a hands‑on AI solutions architect and technical delivery leader with over 15 years of experience designing and delivering AI systems, web applications, platforms and products end‑to‑end.

I currently work as an Architect at Qentelli, helping teams turn business goals into practical software architectures and working closely with engineering to make sure we actually ship. Over the years I’ve moved from Software Developer to Technical Lead, Associate Architect and now Architect, staying close to the code while also owning solution design and delivery.

My experience spans high‑traffic enterprise products, manufacturing platforms like Factory Bucket, and large web applications with thousands of requests per second and millions of users. I’ve led teams across India, Canada and the US, partnering with stakeholders to shape requirements, design architecture, and guide implementation.

I enjoy roles where I can be both technical and customer‑focused: understanding user pain, translating it into clear requirements, and then leading the team to build something reliable and maintainable.

Experience

A brief snapshot of my path

This isn’t a full CV, but it should give you a feel for the kind of problems I’ve worked on and how I think about AI in real products.

  • Architect – Qentelli
    2022 – Present · Texas, US

    Leading architecture and technical delivery for complex web applications and products, combining solution design with project management and stakeholder communication.

    • ✓Ownership of solution requirements and technical roadmap
    • ✓Guiding teams across front‑end, back‑end and integrations
  • Associate Architect & Technical Lead – Qentelli
    2019 – 2022 · Hyderabad, India

    Focused on front‑end and back‑end development, POCs and MVPs, and hiring the right people with the right skill sets for the team.

    • ✓Hands‑on development plus mentoring engineers
    • ✓Participating in solutioning and architectural decisions
  • Sr. Software Engineer – Pro Start Me Technology
    2017 – 2019 · Surat, India

    Built a Canada‑based manufacturing product with Vue.js and Laravel, working directly with clients and owning both front‑end and back‑end responsibilities.

  • Software Engineer – Crest Infosystems
    2015 – 2017 · Surat, India

    Worked on a large US‑based appraisal and banking platform serving millions of users and thousands of requests per second, handling multiple integrations and day‑to‑day client communication.

  • Software Developer – Orange Softworld
    2012 – 2014 · Ahmedabad, India

    Started as a fresher and grew into a full‑stack developer delivering websites, Android applications and Node.js‑based projects, including early work in game development.

What I look for

The collaborations that excite me

If you recognise yourself in this list, there’s a good chance we’ll work well together.

I’m a good fit when you want a partner who can combine architecture, coding and project management, and who is comfortable owning outcomes from idea to production.

I enjoy collaborating with founders, product leaders and engineering managers who value clear communication, honest feedback and a focus on delivery.

Skills & certifications

What I bring to a team

A mix of architecture, hands‑on development and leadership experience built over more than a decade in web and product engineering.

Technical skills

  • ✓Web & product engineering: end‑to‑end web application design and delivery
  • ✓Front‑end: React, ExtJS, Vue.js, modern JavaScript, responsive UI, UX‑aware development
  • ✓Back‑end: Laravel (PHP), Node.js, REST APIs, integrations
  • ✓Data & platforms: MySQL, MongoDB, high‑traffic, highly available systems
  • ✓Ways of working: agile delivery, sprint planning, product planning, team building

Generative AI skills (implementation layer)

  • ✓LLM integration: building applications on top of OpenAI / Claude / Gemini, prompt‑engineering patterns (zero‑shot, few‑shot, chain‑of‑thought), structured JSON output and function calling
  • ✓RAG (retrieval‑augmented generation): embeddings pipelines, vector databases (Pinecone, Weaviate, FAISS), chunking strategies, context ranking and hybrid search
  • ✓GenAI architecture: stateless vs stateful LLM apps, memory handling, token and latency optimisation, cost‑aware design
  • ✓Responsible AI: hallucination mitigation, guardrails, prompt‑injection protection and data‑privacy‑aware design

Agentic AI skills (advanced layer)

  • ✓Agent frameworks: experience with LangChain, LangGraph, CrewAI and AutoGen to design multi‑step, tool‑using agents
  • ✓End‑to‑end design: API‑gateway integration, microservices with an LLM layer, event‑driven AI workflows and CI/CD for AI‑powered services
  • ✓Deployment & ops: shipping AI apps to Vercel / AWS / Azure, containerising with Docker and building serverless AI APIs
  • ✓Observability: LLM logging, prompt tracking, evaluation pipelines and feedback loops
  • ✓Multi‑agent & context graphs: role‑based agents, planner–executor patterns, reflection loops, memory graphs, semantic routing and GraphRAG / knowledge graph integration
  • ✓Autonomous workflow design: task decomposition, self‑evaluation loops, decision trees and tool orchestration

I’m fluent in Gujarati, Hindi and English, which helps when working with distributed teams and clients.

Explore my projects & AI work →