11/11/2026 updated
100 % available
AI Engineer | Production GenAI, RAG Systems & AI Agents | Humenium.AI
Noida, India
Only remote
B.Tech AI & Data Science, Maharaja Agrasen Institute of Technology, Delhi, 2021-2025, CGPA 8.8About me
AI Engineer at DataAlchemy.AI with 2 years building production Generative AI systems. My main work is Humenium.AI (sso.humenium.ai), a live multi-channel notification orchestration platform built with LLM workflows, RAG pipelines and FastAPI backends.
AI AgentsRAGLlamaIndexLangChainApplication Programming Interfaces (APIs)Python (Programming Language)PostgreSQLOpenCVRedisTensorFlowAzure Machine LearningSQL DatabasesChatbotsPyTorchLarge Language Models
Generative AI & Large Language Models
Expertise in Generative AI, LLMs, and RAG-based architectures, including the development of AI-driven applications such as medical chatbots, virtual avatar systems, and automated documentation generators using LangChain, LlamaIndex, and Hugging Face.
Machine Learning & Deep Learning
Proficiency in Machine Learning and Deep Learning using frameworks such as PyTorch, TensorFlow, and Scikit-learn, with hands-on experience in model fine-tuning, inference optimization, and deployment of production-ready AI systems.
API Development & Microservices Architecture
Strong background in building scalable FastAPI-based microservices and event-driven architectures, integrating GitHub APIs, Apache Kafka, and webhook services to enable real-time communication and high-concurrency backend systems.
Python & C++ Programming
Advanced programming skills in Python and C++, applied across AI/ML pipelines, backend services, and automation workflows.
SQL & Database Management
Experience with SQL and SQLite for backend data management, including appointment scheduling, EHR management, and structured data storage.
Docker & Containerization
Practical experience with Docker for containerizing AI applications and microservices, ensuring consistent and scalable deployment environments.
NLP & Text Summarization
Application of NLP-based summarization models to automatically generate structured MARKDOWN documentation and module-level explanations from large codebases.
Computer Vision & Avatar Synthesis
Implementation of text-to-image diffusion models and facial landmark detection for lifelike 3D avatar synthesis, achieving over 92% identity accuracy and 90%+ lip-sync accuracy using Wav2Lip and SadTalker.
Production proof: my main work is Humenium.AI (https://sso.humenium.ai), a live multi-channel notification platform across Email, SMS, WhatsApp, Slack and Voice, with high-throughput queuing, AI templating, and real-time delivery observability. My GitHub documentation generator reaches 95% accuracy and cuts manual effort by 70%; my medical chatbot has handled 1,000+ patient interactions; my avatar pipeline exceeds 90% lip-sync accuracy. Stack: Python (3 years); Generative AI, LLMs, RAG, LangChain, LangGraph, LlamaIndex, FastAPI, Docker, SQL (2 years); PyTorch, TensorFlow, Hugging Face, pgvector, Pinecone, ChromaDB, Qdrant, Weaviate, FAISS, Redis, Kafka, Celery, PostgreSQL, Kubernetes, AWS Bedrock, Whisper, OpenCV (1 to 2 years). Available now for remote work worldwide, $45 per hour introductory rate.
Experience
Professional experience0 - 5 years
Freelancer since2026
Languages
EnglishFluentHindiNative speaker
Project history
Developed an automated GitHub code documentation generator using Python and FastAPI, converting repository structures into structured technical documentation with 95% accuracy and reducing manual documentation time by over 70%. Built a full-stack AI-driven mock interview platform enabling users to practice technical interviews with real-time feedback, integrated question generation, and performance analytics to improve user readiness by an estimated 60%. Architected modular microservices for event-driven automation and API integrations, ensuring scalable communication between GitHub webhook services and documentation pipelines while maintaining sub-second response latency. My main product work is Humenium.AI (https://sso.humenium.ai), a multi-channel notification orchestration platform live in production, routing Email, SMS, WhatsApp, Slack and Voice notifications with high-throughput queuing, AI templating, and real-time delivery observability.
Pioneered an AI-powered documentation generation system that automated extraction and summarization of GitHub repository codebases, reaching 95% accuracy and cutting manual effort by 70%. Programmed an event-driven pipeline leveraging Git APIs and NLP-based summarization models to automatically generate structured MARKDOWN files and module-level explanations for over 500 code commits. Integrated Webhooks, Polling, and Apache Kafka as a distributed event-driven queue, enabling real-time communication across repository workflows and backend microservices.
Humenium.AI (https://sso.humenium.ai) is my main and current work: a multi-channel notification orchestration platform live in production at DataAlchemy.AI. I build routing and delivery across Email, SMS, WhatsApp, Slack and Voice, with high-throughput queuing, AI templating, and real-time delivery observability in a multi-tenant system. Under the hood it runs LLM-powered workflow orchestration with FastAPI microservices, a RAG pipeline for message context using LangChain and pgvector, and Docker-based deployment with monitoring.