Door categorieën bladeren
Ontdekken
Fiverr Pro
Nederlands
$
USD
Maharana Sarkar
@maharana_sarkar
AI and Backend Engineer RAG, LLM, Agents
India
Engels
Over mij
I am an AI & Backend Engineer specializing in Python, FastAPI, Machine Learning, NLP, LLMs and Conversational AI. I build production-ready AI solutions including Rasa chatbots, WhatsApp bots, recommendation systems, vector database & embeddings solutions, AI Agents, and RAG pipelines. I develop scalable backend APIs, deploy using Docker & AWS, and work with MongoDB, Redis, CI/CD, and cloud infrastructure. I focus on clean architecture, performance, and business-ready AI systems.
Skills
Maharana Sarkar
offline •
Gemiddelde reactietijd: 1 uur
Bekijk mijn diensten
AI chatbot development
I will build a production ready ai chatbot using python, rasa or fastapi, docker
AI mobiele apps
I will ai web development, ai application development, ai chatbot and ai agent
Portfolio
Werkervaring
AI Engineer
Beelabs Techknow Care • Fulltime
Mar 2024 - Aug 2025 • 1 yr 5 mos
Designed and developed a job recommendation system microservice using FastAPI, vector embeddings, and PostgreSQL (pgvector), improving candidate–job matching accuracy by ~30%. Built and trained custom embedding models for job and candidate data, integrating FAISS for large-scale, low-latency similarity search. Implemented agentic AI workflows to orchestrate multi-step reasoning for job matching, user intent understanding, and recommendation refinement. Developed and deployed fine-tuned transformer models using LoRA/QLoRA for domain-specific resume understanding and skill inference. Containerized services using Docker and deployed on AWS EC2 with CI/CD pipelines for automated builds and scalable deployment. Built analytics microservices to track chatbot interactions and expose APIs for real-time dashboards and operational monitoring.
AI Engineer
Synthetico AI • Fulltime
Aug 2023 - Mar 2024 • 7 mos
Developed a conversational AI chatbot using Python, NLP, Rasa, and WhatsApp Cloud API, improving appointment scheduling and lab test bookings by ~15%. Implemented location-based search optimization using the Haversine formula to enable faster doctor discovery and matching. Built a retrieval-based embedding pipeline incorporating OCR and image pattern recognition for automated data extraction from images. Applied agent-based reasoning patterns to route user queries dynamically across retrieval, classification, and response-generation modules.
