Imran M

AI Engineer @ Duke Energy | Agentic AI | Data Engineering

I'm an AI Engineer specializing in Agentic AI, RAG pipelines, and cloud-native AI platforms. Currently at Duke Energy, previously PHX Agile and Wells Fargo. With 6+ years building production AI systems using AWS Bedrock, LangGraph, FastAPI, and modern AI observability.

Hire Me My Portfolio

- DARK LIGHT

About me

Personal Info

AI Engineer with 6+ years of experience designing and deploying enterprise AI platforms, Agentic AI systems, RAG pipelines, and cloud native applications. Experienced building production AI solutions using AWS Bedrock, Anthropic Claude, OpenAI API, FastAPI, LangGraph, Python, REST APIs, Docker, Kubernetes, Redis, PostgreSQL, Pinecone, Weaviate, Milvus, FAISS, MLflow, LangSmith, Arize, and modern AI observability practices. Strong background in solution architecture, backend engineering, LLM evaluation, AI governance, and customer facing AI applications.

First Name : Imran
Last Name : M
Role : AI Engineer
Focus : Agentic AI & RAG

Phone : 925-315-5445
Email : Imrangenai08@gmail.com
Company : Duke Energy
Experience : 6+ Years

Download Resume My Portfolio

15+

Project Completed
6+

Years Experience
3+

Happy Customers
5+

Key Highlights

My Skills

Personal Skill

Core skills across Generative AI, Agentic AI, multi-agent systems, RAG, cloud-native backends, vector databases, and AI observability — focused on shipping reliable production AI.

Skills Matrix

Generative AI	Generative AI Agentic AI Multi Agent Systems LangGraph LangChain RAG Prompt Engineering Prompt Versioning Context Engineering MCP
LLMs & APIs	OpenAI API Anthropic Claude AWS Bedrock FastAPI Python Async Python REST APIs Streaming APIs
Data & Vectors	PostgreSQL Redis Pinecone Weaviate Milvus FAISS Vector Databases Semantic Search DynamoDB Athena

Skills Matrix (cont.)

Cloud & Infra	AWS Lambda Step Functions S3 Docker Kubernetes Azure Databricks Event Driven Architecture CI/CD
Observability	MLflow LangSmith Arize AI Observability LLM Evaluation AI Governance Guardrails
Architecture	Solution Architecture Backend Engineering Microservices Cloud Native Production AI Systems

Communication Skill

Communicating (whether by pen, mouth, etc.) in a way that others grasp, Absorbing, sharing, and understanding information presented, Respecting others’ points of view through engagement and interest. Using relevant knowledge, know-how, and skills to explain and clarify thoughts and ideas. Listening to others when they communicate, asking questions to better understand.

English
Hindi
Bengali

Leadership Skill

Partner with stakeholders to deliver production AI capabilities — setting technical direction for Agentic AI workflows, RAG systems, and cloud-native microservices with clear communication and ownership.

Communication
Positivity
Creativity
Responsibility
Giving & Receiving Feedback

My Experience

Personal Experience

I am a self-starter with strong interpersonal skills. I work efficiently both as an individual contributor and with cross-functional teams. I seek new challenges and design creative solutions for enterprise AI problems — with a focus on character, values, vision, and action.

Agentic AI Engineer – Duke Energy
Jan 2025 – Present
Designed and deployed production customer-facing Agentic AI applications using AWS Bedrock, Anthropic Claude, OpenAI API, FastAPI, and REST APIs. Architected cloud-native AI services with Async Python, AWS Lambda, Step Functions, DynamoDB, Athena, S3, and event-driven orchestration. Built production RAG pipelines with FAISS, Pinecone, vector embeddings, semantic search, contextual retrieval, and intelligent chunking. Integrated enterprise knowledge sources through MCP connectors and secure APIs; built LangGraph multi-agent workflows for planning, retrieval, tool execution, and response synthesis. Implemented AI governance, guardrails, prompt versioning, LLM-as-a-Judge evaluation, and observability with MLflow, LangSmith, and Arize.
Technology used:
AI – AWS Bedrock, Anthropic Claude, OpenAI API, LangGraph, RAG, MCP
Backend – FastAPI, Async Python, REST & Streaming APIs
Data – FAISS, Pinecone, PostgreSQL, Redis, DynamoDB, Athena, S3
Infra – Docker, Kubernetes, AWS Lambda, Step Functions
Observability – MLflow, LangSmith, Arize
Project: Diagnostics (GenAI Contact Center Intelligence) – Duke Energy
Jan 2025 – Present
Designed AI-powered transcript intelligence that analyzes customer-agent conversations, generates summaries, extracts actionable insights, and creates structured knowledge assets for downstream AI training and automation. Built scalable ingestion and preprocessing pipelines for transcripts, enabling knowledge extraction and semantic search. Implemented RAG by embedding conversation transcripts for contextual retrieval, with intelligent chunking for large call transcripts to optimize context, accuracy, and token use. Built summarization pipelines enabling next-generation conversational AI agents for voice and chat support.
Focus areas:
Transcript Intelligence, RAG, Vector Embeddings, Semantic Search, Contact Center GenAI

AI Data Engineer – PHX Agile
Jan 2024 – Mar 2025
Designed LangGraph-based multi-agent architectures using Supervisor, Query, Analysis, and Reporting agents. Built FastAPI services exposing AI capabilities through REST APIs and event-driven agent orchestration with Async Python and cloud-native services. Applied prompt engineering, context engineering, guardrails, and LLM evaluation to improve reliability. Built reusable AI workflows supporting enterprise analytics and decision support.
Technology used:
AI – LangGraph, Multi-Agent Systems, Prompt Engineering, Guardrails
Backend – FastAPI, Async Python, REST APIs
Architecture – Event-Driven Agent Orchestration
Associate Software Engineer – Wells Fargo
May 2020 – Dec 2023
Developed enterprise ETL pipelines and backend services supporting critical financial applications for more than 100 clients. Led migration of DataStage workloads and Oracle databases to AWS, improving scalability and deployment efficiency. Built REST-based integrations, automated CI/CD pipelines, and resolved production incidents while maintaining SLA compliance.
Technology used:
Data – ETL, DataStage, Oracle, AWS
Backend – REST Integrations, CI/CD
Ops – Production Incident Management, SLA Compliance

My Achivements & Certificates

Achivements Details

Production Agentic AI Platforms – Duke Energy
2025 – Present
Designed and shipped customer-facing Agentic AI applications with AWS Bedrock, Anthropic Claude, OpenAI API, FastAPI, and LangGraph multi-agent workflows — including planning, retrieval, tool execution, and response synthesis for enterprise use cases.
GenAI Contact Center Intelligence – Diagnostics Project
2025 – Present
Built transcript intelligence solutions that turn customer conversations into searchable knowledge assets — with RAG, vector embeddings, intelligent chunking, summarization, and insight extraction for voice and chat AI agents.
AI Governance & Observability – Enterprise AI
2024 – Present
Implemented content safety guardrails, prompt versioning, and automated LLM evaluation (ground-truth datasets and LLM-as-a-Judge), plus observability with MLflow, LangSmith, and Arize to track response quality, latency, and reliability.
Multi-Agent Analytics Workflows – PHX Agile
2024 – 2025
Delivered reusable LangGraph multi-agent architectures (Supervisor, Query, Analysis, Reporting) and FastAPI AI services supporting enterprise analytics and decision support.
Enterprise Data Migration – Wells Fargo
2020 – 2023
Led migration of DataStage workloads and Oracle databases to AWS while delivering ETL pipelines and backend services for critical financial applications serving 100+ clients, with CI/CD and SLA-focused operations.