Siddharth Singh Portfolio

Summary

I am an AI-focused technology researcher with 5+ years of experience building intelligent software solutions across Generative AI, Agentic AI, semantic search, and backend platform engineering. My work includes enterprise-scale Retrieval-Augmented Generation (RAG), document intelligence, knowledge retrieval, and FastAPI-based services.

At Tata Consultancy Services, I contribute to a custom Semantic Knowledge Engine, supporting document ingestion, parsing, chunking, indexing, embedding generation, metadata filtering, and context-aware retrieval. I evaluate retrieval quality through similarity scoring, ranking, and comparative analysis, while developing APIs and database integrations for enterprise AI platforms.

My personal projects explore multi-agent AI systems, including an autonomous coding assistant using CrewAI and Docker, a deep-research orchestrator using asynchronous agents, and an investment research assistant coordinating specialized agents. These projects reflect my interest in turning LLM capabilities into useful, reliable applications.

Earlier in my career, I worked on the Patent Analytics Tool and data-driven IT operations. I helped analyze 65,000+ monthly alerts for a retail client and developed Power BI reporting that reduced analysis time from hours to minutes.

Experience

Researcher

Tata Consultancy Services

May 2024 - Present

Designed and developed enterprise AI solutions focused on intelligent search, knowledge retrieval, and AI-powered document interactions using Large Language Models (LLMs) and Retrieval-Augmented Generation (RAG).
Contributed to the development of a custom Semantic Knowledge Engine (SKE), implementing chunking strategies, embedding-based indexing, vector retrieval, and metadata-driven search capabilities for enterprise AI applications.
Contributed to enterprise document intelligence initiatives involving parser orchestration, OCR-enabled extraction, hierarchy-aware content processing, and standardized document modeling to improve downstream retrieval accuracy and answer grounding.
Developed and enhanced FastAPI-based backend services, REST APIs, database integrations, and platform capabilities supporting enterprise AI applications and knowledge platforms.
Evaluated retrieval effectiveness and search quality using similarity scoring, ranking metrics, and comparative analysis methodologies to improve the relevance of enterprise knowledge retrieval systems.
Contributed to platform modernization initiatives involving PostgreSQL search capabilities, vector database integrations, API enhancements, testing, and technical documentation.
Collaborated on architecture discussions, feature development, debugging, deployment support, and engineering best practices to ensure scalability, maintainability, and reliability.
Previously contributed to the Patent Analytics Tool (PAT), enhancing patent search, classification recommendations, reporting capabilities, and supporting application compliance, security assessments, and deployments.
Authored architecture documentation, technical guides, and knowledge-transfer material to support onboarding and long-term maintainability.

System Engineer

Tata Consultancy Services

Aug 2021 - Apr 2024

Supported infrastructure monitoring and operations using SolarWinds and NewRelic, helping maintain system stability and operational readiness.
Contributed to a data-driven transformation initiative for a retail client by analyzing 65,000+ monthly alerts to improve IT operational efficiency.
Drove adoption of Power BI for alert analytics and reporting, replacing manual workflows and reducing analysis time from hours to minutes.
Built and maintained a centralized data repository integrated with Power BI, improving data accuracy and reducing repetitive data-cleaning effort.
Developed interactive operational dashboards, including a Starbucks dashboard, to provide actionable insights into performance trends and alert patterns.