I'm a GenAI-focused Computer Science undergraduate with hands-on experience building production-ready RAG (Retrieval-Augmented Generation) applications using FastAPI, LangChain, FAISS, and modern LLM APIs. My engineering approach blends backend system design with applied AI/ML to ship tools that are fast, contextual, and genuinely useful.
I've built and deployed full-stack GenAI products end-to-end — from designing semantic retrieval pipelines and REST APIs to shipping interactive React and Streamlit frontends. I care about clean architecture, low-latency retrieval, and building AI systems that hold up in production, not just in notebooks.
Engineer:
name: "Pushkar Chhokar"
role: "GenAI Engineer | RAG Systems | Full Stack Developer"
focus: ["Retrieval-Augmented Generation", "LLM Backend Engineering", "Agentic AI", "Full Stack Development"]
philosophy: "Ship fast, retrieve smart, engineer with intent."🎯 Open To
| Internships | GenAI / ML Roles | Open Source | Research Collaboration |
|---|---|---|---|
| ✅ | ✅ | ✅ | ✅ |
Languages
Frontend
Backend & Databases
Cloud, DevOps & Tooling
AI / ML & GenAI Frameworks
| Domain | Proficiency | Details |
|---|---|---|
| Retrieval-Augmented Generation (RAG) | ⭐⭐⭐⭐⭐ | Semantic retrieval pipelines, contextual Q&A, custom RAG architectures |
| Generative AI & LLMs | ⭐⭐⭐⭐⭐ | OpenAI GPT, Anthropic Claude, Meta Llama, Google Gemini via OpenRouter API |
| Vector Databases & Retrieval | ⭐⭐⭐⭐ | FAISS, semantic search, embedding pipelines (HuggingFace, SentenceTransformers) |
| Agentic AI & Orchestration | ⭐⭐⭐⭐ | LangChain, Model Context Protocol (MCP), Anthropic-certified Agent Skills |
| Prompt Engineering & Tokenization | ⭐⭐⭐⭐ | Structured prompting, context-window optimization, response shaping |
| Currently Upskilling | ⭐⭐⭐ | PyTorch, LLM Fine-Tuning (LoRA), Cloud AI Deployment (Azure/AWS) |
🔹 AI Learning Copilot — RAG-Powered Learning Platform
A full-stack RAG-powered learning platform enabling contextual PDF Q&A, quiz generation, and flashcard creation across multi-document knowledge bases.
| Aspect | Details |
|---|---|
| Stack | Python, FastAPI, LangChain, FAISS, OpenRouter, SQLite, Streamlit |
| Scale | Handles documents up to 80 pages across knowledge bases of up to 10 PDFs |
| Performance | Sub-2-second semantic retrieval latency via FAISS vector indexing |
| Security | SQLite-based analytics with scoped API access |
| Impact | Adaptive quiz generation, progress tracking, and weak-topic detection to improve retention |
| Repository | View Repository · Live Demo |
Engineered a semantic retrieval pipeline using HuggingFace all-MiniLM-L6-v2 embeddings paired with FAISS indexing, and developed 8+ REST API endpoints powering a personalized learning engine with adaptive analytics.
🔹 AI Resume Analyzer — GenAI Career Intelligence Platform
A production-deployed AI-powered resume analysis platform generating ATS compatibility scores, domain classification, and personalized career recommendations.
| Aspect | Details |
|---|---|
| Stack | Python, FastAPI, React, LangChain, OpenRouter, SentenceTransformers |
| Scale | 6+ REST API endpoints covering analysis, Q&A, and classification |
| Performance | Custom RAG pipeline with cosine similarity search — no external vector DB required |
| Security | Scoped API endpoints deployed via Vercel with isolated request handling |
| Impact | Real-time ATS scoring, AI chat, and interview preparation in an interactive UI |
| Repository | View Repository · Live Demo |
Implemented a custom RAG pipeline using SentenceTransformers (all-MiniLM-L6-v2) for contextual resume Q&A, and shipped an interactive React frontend supporting ATS scoring, AI chat, and interview prep — fully deployed on Vercel.
🔹 DevMind — Multi-Agent AI Code Review Platform
An AI-powered repository analysis platform that automatically clones Git repositories, scans source code, generates semantic embeddings, and performs multi-agent code reviews to deliver actionable engineering insights.
| Aspect | Details |
|---|---|
| Stack | Python, FastAPI, React, SQLAlchemy, PostgreSQL, LangChain, FAISS, GitPython |
| Architecture | Planner, Retriever, Reviewer, Critic, and Report Generation pipeline |
| Performance | Semantic repository analysis with intelligent chunking and vector-based retrieval |
| Scalability | Modular service-oriented backend with asynchronous job processing |
| Impact | Automated code quality review, architecture analysis, security observations, and maintainability reporting |
| Repository | View Repository |
Designed and implemented a production-oriented multi-agent AI system that clones Git repositories, extracts source code metadata, generates vector embeddings, and orchestrates specialized AI agents to analyze code quality, architecture, and engineering best practices through an end-to-end automated review pipeline.
June 2026
Completed a virtual job simulation building AI-powered financial intelligence tooling.
- Built an AI-powered financial chatbot using LLMs for financial document Q&A
- Extracted, processed, and analyzed structured financial data for intelligent retrieval
- Applied prompt engineering, generative AI, and NLP to real-world business use cases
Generative AI LLMs NLP Prompt Engineering
June 2026
Completed a virtual job simulation focused on AI-assisted proposal generation for higher education institutions.
- Created AI-assisted hypothesis proposals using generative AI
- Refined proposal content through iterative discovery and stakeholder feedback
- Applied prompt engineering and value-selling strategies to improve client outcomes
Generative AI Prompt Engineering Business Analysis
June 2026
Completed a simulated consulting engagement in data analytics and forensic technology.
- Performed data analysis and forensic technology tasks in a consulting environment
- Analyzed datasets to identify trends, anomalies, and actionable business insights
- Strengthened data interpretation and client-oriented decision-making skills
Data Analytics Forensic Technology Critical Thinking
| Recognition | Details |
|---|---|
| 📄 Registered Copyright | Authored and obtained copyright registration for "Copyright Issues in Generative AI" — an analysis of intellectual property challenges in LLM-generated content (2026) |
| 🚀 Production-Deployed GenAI Products | Shipped two full-stack RAG applications live on Streamlit and Vercel |
| 🧩 Anthropic-Certified Agent Skills | Completed certification in Agent Skills and Model Context Protocol |
Anthropic
Google Cloud
IBM
Learning:
- PyTorch for deep learning workflows
- LLM Fine-Tuning with LoRA
- Cloud AI Deployment (Azure / AWS)
Building:
- Production-grade RAG applications
- Agentic AI workflows using MCP and LangChain
Exploring:
- Advanced embedding strategies for semantic search
- Multi-agent orchestration systems
Open To:
- GenAI / ML Engineering Internships
- AI Backend Engineering Roles
- Open Source Collaboration
"Retrieval is only as good as the context you build around it — engineer with intent."