AI Engineer with 1+ years in RAG, Agentic Workflows & MLOps
AI is analyzing your overall score…
Identifying your key strengths…
Evaluating your skill match against the job requirements…
Assessing your cultural and operational fit
Self-taught AI Engineer with 1+ years of hands-on experience building production-grade RAG applications and agentic workflows. Strong full-stack AI capabilities including data pipelines, LLM-powered systems, FastAPI backend development, AWS deployment, and Supabase integration. Proven through hackathon prototypes and personal deep-dive LLM projects. Passionate about creating scalable, autonomous AI solutions and efficient MLOps practices.
SR UNIVERSITY
B-Tech · Mechanical Engineering
August 1, 2021 – June 30, 2025
ENVYNEX
AI Engineer
February 1, 2025 – Present
Hyderābād, Telangana, India
Public Eye Agent Prototype – Intelligent Procurement Audit & Fraud Detection System
December 1, 2025 – December 31, 2025
Engineered an agentic AI prototype for detecting fraud and anomalies in public procurement contracts, combining rule-based logic and machine learning. Implemented over 50 deterministic rules (e.g., sole-source awards, spending spikes) with unsupervised anomaly detection using Isolation Forest. Built a full-stack solution: FastAPI backend for API endpoints, Streamlit for interactive UI, Google ADK and Gemini for conversational agent explanations and tool routing. Enabled flexible CSV data ingestion, dynamic column mapping, batch processing for large datasets (e.g., 1M+ rows), session management, and hybrid risk scoring. Showcased agentic capabilities via natural language Q&A for audit insights, demonstrating potential for governance and compliance applications. Competed in a major agentic AI hackathon, highlighting rapid prototyping and production-viable design.
View ProjectLLM from Scratch – End-to-End LLM Training Pipeline
January 1, 2024 – December 31, 2025
Developed a comprehensive Jupyter-based pipeline for building, training, and fine-tuning small GPT-like LLMs from foundational concepts. Handled tokenization with tiktoken (BPE), dataset preparation from public sources (e.g., instruction and spam datasets), and model definition in PyTorch. Integrated GPT-2 weight loading, supervised fine-tuning (SFT), training loops with progress tracking (tqdm), checkpointing, and local inference via Ollama API. Explored core LLM workflows including data preprocessing, evaluation metrics, and troubleshooting, enhancing expertise in transformer-based models. Applied learnings to inform RAG optimizations and agentic system designs in professional work, bridging theory to practical AI deployment.
View ProjectAgentathon 2025 Participant (Focused on Agentic AI Innovations)
Google Developer Groups Hyderabad
January 1, 2025 – Present
Cultural Fit Analysis
The candidate demonstrates a strong passion for AI, evidenced by self-study, hackathon participation, and personal projects. The focus on building 'production-grade' and 'scalable, autonomous AI solutions' aligns well with an innovative and results-oriented culture. The breadth of skills across AI/ML, backend, data, and cloud indicates adaptability and a willingness to tackle diverse challenges. However, the limited professional experience (1 year) and non-traditional educational background might require additional mentorship in a structured corporate environment.
Soft Skills & Operational Fit
The candidate's project descriptions indicate a proactive and self-driven approach to learning and problem-solving, particularly in transitioning from Mechanical Engineering to AI. Participation in hackathons suggests a collaborative and competitive spirit. The detailed descriptions of production-grade systems imply an understanding of operational requirements and scalability.