Aluguri Vinay | Gen AI Engineer | Agentic AI Engineer

01 //

System Overview

AI Engineer with hands-on experience designing and building GenAI systems using Python, LangChain, and Retrieval-Augmented Generation (RAG) pipelines.

Skilled in prompt engineering, embeddings, vector databases (ChromaDB, FAISS, Pinecone), and agent-based frameworks (CrewAI) for context-aware, production-grade AI solutions. Experienced building FastAPI backend services and REST APIs, integrating OpenAI, Claude, Gemini, Groq, and deploying scalable architectures to AWS.

Engine Designation Gen AI Engineer | Agentic AI | LLM

Primary Node Proddatur, Andhra Pradesh, India

Secure Pipeline vinayaluguru@gmail.com

Direct API Dial +91 9100129701

Professional Index LinkedIn Profile

02 //

Execution Pipeline

AI Engineer (Remote)

Turing (Remote)

MAR 2025 – JUL 2026 · 1 YR 5 MOS

Owned end-to-end delivery of RAG and multi-agent AI systems, from architecture design through production deployment on AWS.
Led integration of multiple LLM providers (OpenAI, Claude, Gemini, Groq) into production workflows, ensuring reliable, context-aware responses at scale.
Drove adoption of evaluation practices across projects—tracking retrieval accuracy, groundedness, latency, and hallucination rate to guide iteration.
Integrated agentic AI frameworks and automation tools (LangChain, CrewAI, MCP, n8n) to build scalable backend services and automated workflows.
Collaborated with cross-functional stakeholders to translate business requirements into scalable AI solutions and mentored on best practices for prompt engineering, agent orchestration, and CI/CD deployment.

LLM S2 Annotator (OpenClaw CUA Trajectory Specialist)

Turing (Remote)

NOV 2024 – JAN 2025 · 3 MOS

Created Computer Use Agent (CUA) trajectories from natural-language instructions to align multi-step agent behavior.
Evaluated AI-generated outputs for reasoning quality, factual accuracy, safety compliance, and strict instruction adherence.
Performed prompt evaluation, pairwise (SxS) response ranking, and automated hallucination detection across complex outputs.
Investigated workflow and execution failures through systematic trace analysis and tool-call validation.
Validated multi-turn reasoning workflows, execution consistency, and context retention across multi-step execution paths.
Collaborated with cross-functional AI research teams to develop evaluation guidelines and improve model trajectory alignment datasets.

LLM Agentic Trainer

Turing (Remote)

AUG 2024 – OCT 2024 · 3 MOS

Trained AI/LLM models using Python, SQL, Git, JSON, and TAU tools across consecutive AI engineering workflows.
Designed and developed agentic workflows, JSON-based task structures, and tool-integration setups for multi-step execution.
Applied prompt engineering (Zero-shot, Few-shot, Chain-of-Thought, ReAct) and RLHF-based evaluation methodologies to improve model reasoning, trajectory accuracy, and output quality.
Evaluated multi-turn model trajectories to ensure strict criteria adherence, logical coherence, context retention, and hallucination reduction across agentic tasks.
Developed evaluation guidelines, benchmark prompts, and documentation supporting scalable AI training and alignment datasets.

03 //

Systems & Showcase

RAG Platform // Production [ DEPLOYED ON AWS ]

Multi-Source RAG Platform

Built a production-ready RAG platform supporting PDF, DOCX, CSV, TXT, and website ingestion for contextual Q&A. Developed scalable FastAPI backend microservices for document ingestion, embedding generation, semantic retrieval, and AI workflow execution. Integrated OpenAI, Claude, Gemini, and Groq with CI/CD on AWS EC2.

Python FastAPI LangChain ChromaDB Hugging Face OpenAI AWS EC2

Agentic AI // Orchestration [ DEPLOYED ON AWS ]

Multi-Agent AI Automation Platform

Built a production-ready multi-agent platform using CrewAI to automate research, planning, content generation, and workflow execution. Designed specialized agents (Researcher, Planner, Writer, Reviewer) with role-based task delegation, memory management, and FastAPI REST API orchestration deployed on AWS.

CrewAI Python FastAPI LangChain Tavily OpenAI AWS EC2

Conversational AI [ REALTIME VOICE ]

AI Voice Agent

Built a real-time voice agent integrating speech-to-text, LLMs, and conversational AI. Developed scalable FastAPI backend services for speech processing (Whisper), AI inference (OpenAI, Claude, Gemini), and workflow orchestration with latency and hallucination evaluation.

Python FastAPI Whisper OpenAI Claude Gemini

Automated Research [ DEPLOYED ON AWS ]

AI Research Assistant

Built an AI research assistant that automates web research, summarization, and report generation using LangChain and Tavily Search. Designed n8n automation workflows to orchestrate multi-step research pipelines and exposed query routing via FastAPI REST APIs on AWS EC2.

Python FastAPI LangChain Tavily n8n OpenAI AWS EC2

04 //

Intelligence Matrix

Core LLM & Generative AI

Core LLM Stack & Python 96%

Python, LangChain, SQL, FastAPI REST APIs, Microservices architecture.

Generative AI & Vector DBs 94%

LLM Internals, Prompt Tuning, Embeddings, ChromaDB, FAISS, Pinecone.

Agent Workflows & RAG 95%

CrewAI, MCP Protocol, Semantic Search, Tool Calling, Hallucination Detection.

Frameworks & Cloud Infrastructure

AI Frameworks & Tooling 92%

LangChain, Hugging Face, CrewAI, MCP, n8n, Cursor, Claude Code, Copilot.

Cloud & DevOps (AWS & Docker) 91%

AWS EC2, Docker Containerization, Git, GitHub Actions CI/CD, REST API Microservices.

Evaluation & Engineering Practices 93%

Retrieval Accuracy, Latency Tracking, SxS Ranking, Technical Documentation.

Engineering
Agentic AI Systems

System Overview