Pranjul/ resume

Summary

GenAI engineer and AI product builder with 2+ years shipping production-grade AI systems — Retrieval-Augmented Generation (RAG) pipelines, fine-tuned LLMs, hybrid retrieval, and multi-modal AI applications spanning vision, speech, and OCR — architected end-to-end from data ingestion through deployment. Led engineering for a 14-member team building AI-powered products, pairing deep GenAI/ML engineering with full-stack delivery, while driving measurable business revenue as a startup technical lead.

Experience

  1. NEXT UPGRAD WEB SOLUTIONS

    Feb 2026 – Jul 2026

    GenAI Engineer

    • Independently designed, built, and shipped 5 production-grade AI applications end-to-end — spanning computer vision, document intelligence, LLM fine-tuning, speech-to-text/OCR, and retrieval-augmented generation (full list under Projects).
    • Built a production RAG platform with hybrid (dense + sparse) retrieval, re-ranking, and streaming citation-grounded answers, including a confidence gate that blocks the model from answering on weak evidence.
    • Built an end-to-end LLM fine-tuning platform (QLoRA via Hugging Face Transformers/PEFT/TRL) with live training telemetry and side-by-side base-vs-fine-tuned evaluation.
    • Deployed and operated all five services on Railway/Vercel with Docker, CI/CD, and observability (structured logging, metrics), achieving production-grade reliability across the portfolio.

  2. SCULT INDIA

    2024 – Present

    Senior GenAI Engineer

    • Built and shipped 5+ paid, production-grade web applications and AI-powered products for clients across industries.
    • Led engineering, system architecture, UI/UX, deployment, and client delivery for a cross-functional 14-member team.
    • Designed and developed SCULT Studio CRM, an internal platform for project, client, and operations management.

  3. TechVerse Enclave

    2023 – Present

    Mentor & Tutor

    • Mentor 200+ students across a 500+ member developer community in web development and career growth.
    • Host multiple online workshops to teach followers and juniors practical web development and career skills.

Technical skills

AI/ML & GenAI
LLMs (OpenAI, Claude, Gemini, Mistral, Groq/Llama), RAG Pipelines & Hybrid (Dense+Sparse) Retrieval, LLM Fine-Tuning (LoRA/QLoRA via Hugging Face Transformers/PEFT/TRL), Vector Search & Embeddings, Prompt Engineering, LangChain, MCP (Model Context Protocol), OpenRouter, ONNX Runtime, Computer Vision
Languages
Python, JavaScript, TypeScript, HTML5, CSS3
Backend
FastAPI, Node.js, Express.js, REST APIs, WebSockets
Databases & Vector Stores
PostgreSQL, MongoDB, Supabase, Firebase, Redis, FAISS, Qdrant, pgvector
Frontend
React.js, Next.js, Tailwind CSS, Shadcn UI, Redux, Zustand
Cloud & Tools
AWS, Docker, Vercel, Railway, Git, GitHub, CI/CD

Projects — NextUpgrad Web Solutions (featured AI projects)

  • FaceVisionDemo GitHub

    Browser-only face detection, recognition & liveness-check app (ONNX Runtime Web); FastAPI/PostgreSQL backend stores only vector embeddings, never images.

  • DocuLens AIDemo GitHub

    Groq-LLM document-intelligence platform — extracts, compares & checks compliance on invoices/contracts with schema-validated, deterministic scoring.

  • FineTune StudioDemo GitHub

    End-to-end QLoRA fine-tuning platform (HF Transformers/PEFT/TRL) with live loss streaming and base-vs-fine-tuned evaluation.

  • OCR & Speech WorkspaceDemo GitHub

    Mistral-AI workspace for concurrent page-batched PDF OCR, live speech-to-text, and document-scoped RAG chat with citations.

  • RAG.NextUpgradDemo GitHub

    Production RAG platform with hybrid dense + BM25 retrieval, a hallucination-blocking confidence gate, and automatic multi-provider fallback.

Projects — Independent & client projects

Education

Dr. Virendra Swarup Institute of Computer Studies, Kanpur

2023 – 2026

Bachelor of Computer Applications (BCA) · CGPA: 8.6 / 10