harsh = {
"role" : "AI/ML Engineer @ Logicwind",
"location" : "Surat, Gujarat, India 🇮🇳",
"experience" : "1.5 years in production AI systems",
"education" : "MCA — Gujarat Technological University (CGPA: 8.31)",
"focus" : ["LLM Integration", "RAG Pipelines", "Computer Vision", "AI Microservices"],
"tools" : ["LangChain", "OpenAI", "YOLO", "FastAPI", "Docker", "FAISS"],
"available" : True # Open to freelance & full-time opportunities
}I don't build demos — I build production systems. In the past 1.5 years I've shipped 8+ client AI applications that are live, containerized, and running on cloud infrastructure. My work spans LLM-powered virtual assistants, multi-modal RAG pipelines, computer vision microservices, and OCR systems for specialized domains like medical imaging and legal documents.
|
Production-ready enterprise AI assistant with source-attributed Q&A, Elasticsearch-powered hybrid search, multi-version LLM support, and document ingestion pipelines. Deployed as Dockerized FastAPI + Flask microservices. Stack: |
FastAPI microservice for automated graphology — YOLO + Google Vision API extract personality traits and red flags from handwriting samples. Async job processing via Redis/RQ with webhook notifications. Stack: |
|
Cloud-based system automating floorplan detection, validation, and Vastu compliance analysis using AI + image processing. Rule-driven architectural assessment with cloud microservices. Stack: |
Comprehensive AI pipeline for text and table extraction from medical images — supports JPEG, PNG, and DICOM formats with spatial text correction and medical term normalization. Stack: |
|
Scalable containerized virtual assistant for real-time conversational AI + image generation. Multi-model support via OpenRouter and Replicate with content moderation and session management. Stack: |
PDF Q&A chatbot using Retrieval-Augmented Generation — FAISS persistent vector storage, sentence-transformer embeddings, and Groq-powered LLM inference with a Streamlit UI. Stack: |
| 🏢 Role | 🏗️ Company | 📅 Period |
|---|---|---|
| AI/ML Engineer | Logicwind, Surat | Aug 2024 – Present |
| AI/ML Intern | Logicwind, Surat | Feb 2024 – Jul 2024 |
- 🔹 Built multi-modal RAG system querying video, audio, PDFs & web data using LangChain + Guardrails AI
- 🔹 Engineered production RAG chatbot with FAISS vector storage + sentence-transformer embeddings
- 🔹 Developed AI Website Builder — natural language → responsive HTML/CSS via OpenAI + Streamlit
- 🔹 Built handwriting analysis microservice with YOLO + Google Vision API + async Redis/RQ pipeline
- 🔹 Created medical OCR system supporting DICOM, JPEG, PNG with spatial text correction
- 🔹 Designed legal document scraper integrating OCR + ML pipelines + PostgreSQL storage
| 🎓 Degree | 🏫 University | 📅 Year | ⭐ CGPA |
|---|---|---|---|
| Master of Computer Application | Gujarat Technological University | 2021 – 2023 | 8.31 / 10 |
| Bachelor of Computer Application | Gujarat Technological University | 2018 – 2021 | 7.04 / 10 |
I'm open to freelance projects, full-time roles, and interesting AI/ML collaborations.
