# Deep Gandhi

## Contact
- Location: Pune, Maharashtra, India
- Email: gandhideep91@gmail.com
- LinkedIn: https://www.linkedin.com/in/dgandhi91/
- GitHub: https://github.com/dgandhi91

## Summary

Principal Software Engineer with 12+ years of experience designing and delivering enterprise software and production AI systems. Specialized in Generative AI, Agentic AI, Retrieval-Augmented Generation (RAG), AI evaluation, enterprise search, and cloud-native platforms.

Experienced in leading technical architecture for enterprise AI initiatives, building scalable AI platforms, and translating complex business requirements into production-ready solutions. Passionate about developing reliable, observable, and governable AI systems that improve customer and engineer productivity.

## Professional Experience

### Principal Software Engineer
**Red Hat** | Pune, India

Lead architecture and technical direction for enterprise AI initiatives, collaborating across engineering, product, and leadership teams to deliver production-ready AI capabilities.

#### Enterprise AI Assistant (Ask Red Hat)
Enterprise AI assistant supporting customer self-service and troubleshooting.

- Architected an Agentic RAG platform for enterprise support
- Designed hybrid retrieval and re-ranking pipelines to improve answer quality
- Optimized inference performance through prompt and token management
- Improved response grounding and reduced hallucinations using evaluation-driven improvements
- Guided technical direction across multiple engineering teams

Technologies: LLMs, LangGraph, LangChain, GraphQL, Solr, vLLM, Kubernetes, OpenShift

#### AI Case Summarization
AI-powered summarization for enterprise customer support cases.

- Designed scalable summarization workflows for long-running support cases
- Built evaluation datasets and quality metrics to measure summary accuracy
- Improved summarization quality through prompt engineering and iterative evaluation
- Reduced manual effort for support engineers by automating case analysis

Technologies: LLMs, Prompt Engineering, AI Evaluation

#### Portal Case Management (Conversational Support Agent)
Natural language interface for enterprise case lifecycle management.

- Designed conversational workflows for case creation, updates, retrieval, and management
- Integrated AI agents with enterprise case management systems using secure tool execution
- Implemented intelligent intent routing and multi-step task execution
- Defined evaluation metrics for tool selection, routing accuracy, and task completion

Technologies: Agentic AI, MCP, LangGraph, REST APIs, GraphQL

#### Enterprise AI Evaluation Framework
Designed a standardized evaluation platform covering the complete AI lifecycle.

Capabilities:
- Retrieval evaluation
- Generation evaluation
- Agent evaluation
- Tool evaluation
- Hallucination detection
- Faithfulness and context precision measurement
- LLM-as-a-Judge evaluation
- Production monitoring and tracing
- Feedback analysis and continuous quality improvement

### Software Engineer (Previous)
Designed and developed enterprise web applications and backend services using Java and modern open-source technologies.

- Backend application development
- REST API development
- Microservices
- Database design
- Cloud-native application development

## Technical Skills

### Artificial Intelligence
- Generative AI
- Agentic AI
- Large Language Models (LLMs)
- Retrieval-Augmented Generation (RAG)
- Prompt Engineering
- AI Evaluation
- AI Governance
- LLM-as-a-Judge

### AI Frameworks
- LangGraph
- LangChain
- LlamaIndex
- Model Context Protocol (MCP)

### Programming Languages
- Java
- Python

### Backend Technologies
- Spring Boot
- REST APIs
- GraphQL

### Search & Retrieval
- Solr
- BM25
- Hybrid Search
- Re-ranking
- Embeddings

### Databases
- PostgreSQL
- PGVector
- ChromaDB
- SQLite

### Cloud & Infrastructure
- Kubernetes
- OpenShift
- Docker
- Podman
- GitLab CI/CD
- vLLM

### Observability
- Langfuse
- Evaluation Pipelines
- Distributed Tracing

### AI Models
- IBM Granite
- Llama
- Qwen
- Mistral

## Core Expertise

- Enterprise AI Architecture
- AI Platform Engineering
- Agentic AI Systems
- Conversational AI
- Knowledge Retrieval
- Enterprise Search
- AI Evaluation & Governance
- LLM Infrastructure
- Cloud Native Engineering
- Distributed Systems
- Software Architecture
- Technical Leadership
- Cross-functional Collaboration

## Selected Impact

- Architected production AI systems supporting enterprise customer support
- Established AI evaluation and governance practices for production deployments
- Built scalable retrieval and inference architectures for enterprise AI applications
- Designed conversational AI capabilities for enterprise case management
- Improved AI quality through evaluation-driven development and production monitoring
- Mentored engineers and influenced technical strategy across multiple teams
- Delivered AI capabilities contributing to approximately $14.9M in business impact

## Education

Bachelor of Engineering, Computer Engineering

## Interests

- Enterprise AI
- Agentic AI
- Open Source AI
- Knowledge Systems
- Software Architecture
- Distributed Systems

---

## About This File

This is a machine-readable CV file optimized for AI agents and LLMs.
Full profile: https://www.puneaicollective.org/people/deep

Last updated: 2026-07