Durga Venu Kumar Mutyala
Profile

Python Developer & Generative AI Engineer with 1 year of experience in building LLM-based applications, RAG systems, and agentic workflows. Skilled in Python, LangChain, vector databases, and prompt engineering. Experienced in delivering scalable, intelligent AI solutions with a strong focus on backend development and real-world impact.

Work Experience
Gen AI Engineer | Tech Mahindra
02/2026 – Present
  • •Leading the design, development, testing, and deployment of an enterprise-grade conversational AI chatbot using Retrieval-Augmented Generation (RAG) architecture grounded in client’s proprietary knowledge base.
  • •Architected full RAG pipeline including document ingestion, intelligent chunking, vector store, retrieval optimization, prompt engineering, and evaluation loops to measure response accuracy, faithfulness, and hallucination rates.
  • •Collaborating with cross-functional stakeholders (Product, Business, Engineering) in Agile environment to deliver production-ready Agentic AI features.
    • Gen AI Developer | TechMahindra
    11/2024 – 01/2026
  • •Designed, developed, and deployed a production-scale Gemini 2.5 Flash multi-modal LLM pipeline that processed and structured 500M+ low-quality scanned TIFF documents, achieving 95% extraction accuracy across 9 metadata fields.
  • •Built and productionized a Human-in-the-Loop (HITL) confidence scoring system using LLM token log probabilities, routing only bottom 5% for manual review — reducing human workload by 95%.
  • •Awarded ‘Pat on the Back’ for driving Generative AI innovation and cross-functional collaboration.
    • Skills
    Programming: Python, FastAPI, Flask, Pandas, NumPy
    Agentic AI & LLM: LangGraph, Agentic Workflows, Multi-Agent Orchestration, Prompt Engineering
    RAG and Vector Search: RAG, Qdrant, ChromaDB, FAISS
    Databases: Postgres, Vector DB
    System Design: Distributed Systems, Multi-tiered Architectures, Algorithms & Data Structures
    DevOps & Tools: Docker, Git, GitHub Actions (CI/CD), Automated Testing, Streamlit, Reflex
    Cloud & Methodologies: Azure, GCP, Agile (Scrum/Kanban)
    Other: LLM Evaluation & Observability, Hallucination Mitigation, Human-in-the-Loop Systems
    Education
    B. Tech in Electronics and Communication, Vishnu Institute of Technology
    2020 – 2024
    Projects
  • •- Developed a multi-step LangGraph-powered autonomous agent that analyzes master resume and target Job Description, then rewrites, reorders, and tailors content while strictly grounding every output to source data to ensure zero hallucination.
  • •Delivered highly accurate, traceable, ATS-friendly resumes through end-to-end agentic workflow.
  • •Built an end-to-end Agentic AI application using LLM vision + reasoning that extracts context from LinkedIn hiring screenshots, generates personalized emails in user’s style, and sends via SMTP.
  • •Orchestrated complete agentic flow (perception → reasoning → action) reducing application time from minutes to seconds.
    • Docsify – Documentation to Grounded Conversational Chatbot
  • •Created a scalable system that recursively crawls documentation sites, performs intelligent chunking & embedding into vector stores with incremental updates and deduplication.
  • •Implemented source-attached retrieval ensuring full traceability with page titles and URLs for every response.
    • AWARDS, ACHIEVEMENTS & PUBLICATIONS
  • •1st Place – Microsoft Agentic AI Hackathon 2026 (Microsoft & Tech Mahindra)
  • •1st Place – AI for India Hackathon 2024 (Axamine AI & OpenRag)
  • •Google Cloud Generative AI Skill Boost Program (Multiple Badges)
  • •‘Pat on the Back’ Award – Tech Mahindra, Nov 2025
  • •Open-source Contributor – Flipkart Product Scraper (GitHub)