Dinesh Gaddi profile photo 1 on Tal

Dinesh Gaddi

Data Scientist

Mumbai

14+ years of work experience

Qyrus
Mar 20264 mosPresent

Mumbai, Maharashtra, India

Building an AI-powered code analysis and modernization platform for IBM AS400 systems using RAG architecture. Implemented smart code ingestion, dependency graph generation, and metadata-aware embeddings stored in Milvus for high-accuracy retrieval. Developed LLM-driven APIs to automatically generate business rules, system design, migration strategy, and technical documentation. Focused on scalable, enterprise-grade architecture with visualization, knowledge graphs, and automated document generation.
Quinnox
Jan 20188 yrs 2 mosFeb 2026

Mumbai Metropolitan Region

Delivered innovative AI-powered solutions that bridge the gap between human language and complex technical systems, enabling faster decision-making, improved efficiency, and modernization of legacy processes. Reduced IT ticket routing time by 70%, accelerated documentation creation from weeks to hours, and empowered non-technical teams to access and analyze data effortlessly. Key Achievements & Responsibilities: Developed a suite of AI Agents — TalkToDB, TalkToDocument, TalkToCode, and TalkToKnowledgeGraph — to enable natural language interaction with databases, documents, codebases, and graph databases. Automated IT Ticket Classification (assigned group & priority) using custom rule documents in TalkToDocument, improving routing accuracy and reducing manual errors. Led POCs for AI-Generated Documentation — produced Technical, SRS, and Business documents from legacy (IBM AS400/COBOL) and modern (Java, Python) codebases, extracting workflows, dependencies, and business logic. Built ReAct Agent Pipelines integrating custom tools: Pandas Query Engine — natural language to structured data analysis. Neo4j Tool — Cypher queries and graph summarization. OpenAPI Tool — automated interaction with APIs from OpenAPI specs. Developed Interactive AI Applications using Gradio for an intuitive, user-friendly experience. Core Technologies & Skills: AI Frameworks: LlamaIndex (extensive use for RAG, document ingestion, and tool orchestration). Vector Databases: PGVector, Milvus (semantic search, context retrieval). LLMs: Mistral-7B, Llama-8B, Claude, OpenAI (code reasoning, documentation generation, NLP). Techniques: Prompt Engineering, Retrieval-Augmented Generation (RAG), Multi-Agent Orchestration, Data Pipeline Development. Databases: SQL, Neo4j Graph DB. Frontend/UI: Gradio (interactive AI apps).
Shotformats Digital Productions Pvt Ltd
Sep 20152 yrs 5 mosJan 2018

Mumbai Area, India

Creating wap sites and CMS using codeigniter. Creating Hybrid mobile application.
Indian Institute of Technology, Bombay
Apr 20132 yrs 5 mosAug 2015

Involved in developing websites using very popular CMS Drupal Developed custom modules in Drupal 6 for the website Maintenance of the website Responding to the queries related to the website
Smart Solutions
Jul 201210 mosApr 2013

Virar

Working on LAMP technologies Drupal (CMS). Drupal.org profile: https://www.drupal.org/u/dineshsgaddi GitHub: https://github.com/dineshsg/

Education

Bachelor of Engineering (B.E.), Computer Science

S.D.M.C.E.T

20047 yrs2011

Diploma, Computer Technology

Terna polytechnic

20013 yrs2004

Join Dinesh and 10k+ other professionals on

Tal Boss

Tal is the app where bosses meet engineers & builders directly, without middlemen

Download Tal Boss

Dinesh Gaddi on Tal

Headline

Data Scientist

About

To be a trustworthy person both as a professional and a man

Location

Mumbai

Work experience

Data Science Manager at Qyrus, Mar 2026 - Present · 4 mos, Mumbai, Maharashtra, India, Building an AI-powered code analysis and modernization platform for IBM AS400 systems using RAG architecture. Implemented smart code ingestion, dependency graph generation, and metadata-aware embeddings stored in Milvus for high-accuracy retrieval. Developed LLM-driven APIs to automatically generate business rules, system design, migration strategy, and technical documentation. Focused on scalable, enterprise-grade architecture with visualization, knowledge graphs, and automated document generation.

Data Science Manager at Quinnox, Jan 2018 - Feb 2026 · 8 yrs 2 mos, Mumbai Metropolitan Region, Delivered innovative AI-powered solutions that bridge the gap between human language and complex technical systems, enabling faster decision-making, improved efficiency, and modernization of legacy processes. Reduced IT ticket routing time by 70%, accelerated documentation creation from weeks to hours, and empowered non-technical teams to access and analyze data effortlessly. Key Achievements & Responsibilities: Developed a suite of AI Agents — TalkToDB, TalkToDocument, TalkToCode, and TalkToKnowledgeGraph — to enable natural language interaction with databases, documents, codebases, and graph databases. Automated IT Ticket Classification (assigned group & priority) using custom rule documents in TalkToDocument, improving routing accuracy and reducing manual errors. Led POCs for AI-Generated Documentation — produced Technical, SRS, and Business documents from legacy (IBM AS400/COBOL) and modern (Java, Python) codebases, extracting workflows, dependencies, and business logic. Built ReAct Agent Pipelines integrating custom tools: Pandas Query Engine — natural language to structured data analysis. Neo4j Tool — Cypher queries and graph summarization. OpenAPI Tool — automated interaction with APIs from OpenAPI specs. Developed Interactive AI Applications using Gradio for an intuitive, user-friendly experience. Core Technologies & Skills: AI Frameworks: LlamaIndex (extensive use for RAG, document ingestion, and tool orchestration). Vector Databases: PGVector, Milvus (semantic search, context retrieval). LLMs: Mistral-7B, Llama-8B, Claude, OpenAI (code reasoning, documentation generation, NLP). Techniques: Prompt Engineering, Retrieval-Augmented Generation (RAG), Multi-Agent Orchestration, Data Pipeline Development. Databases: SQL, Neo4j Graph DB. Frontend/UI: Gradio (interactive AI apps).

Software Engineer at Shotformats Digital Productions Pvt Ltd, Sep 2015 - Jan 2018 · 2 yrs 5 mos, Mumbai Area, India, Creating wap sites and CMS using codeigniter. Creating Hybrid mobile application.

Junior Software Engineer at Indian Institute of Technology, Bombay, Apr 2013 - Aug 2015 · 2 yrs 5 mos, Involved in developing websites using very popular CMS Drupal Developed custom modules in Drupal 6 for the website Maintenance of the website Responding to the queries related to the website

Software Trainee at Smart Solutions, Jul 2012 - Apr 2013 · 10 mos, Virar, Working on LAMP technologies Drupal (CMS). Drupal.org profile: https://www.drupal.org/u/dineshsgaddi GitHub: https://github.com/dineshsg/

Education

Bachelor of Engineering (B.E.), Computer Science, S.D.M.C.E.T

Diploma, Computer Technology, Terna polytechnic

Dinesh Gaddi - Data Science Manager in Mumbai | Tal