Yashowardhan Singh Tomar
RAG / LLM Application Engineer | Retrieval Evaluation | Python
Indore, India | Open to remote and relocation | yashtomar10122@gmail.com | +91-97549-50809
linkedin.com/in/yashowardhansinghtomar | github.com/yashowardhansinghtomar
Profile
AI engineer building practical LLM applications with a focus on RAG, document QA, retrieval evaluation, and local-first GenAI systems. Public portfolio includes a RAG evaluation workbench, PDF QA workflows, local Ollama assistants, and LLM evaluation tooling. Background combines applied Python development, LLM behavior evaluation, voice AI integration, and business-facing product experience.
Experience
Jr. AI / Data Science Engineer | Engineer MasterSep 2024 - Jan 2025
- Built and evaluated LLM workflows involving prompt design, structured model review, and failure analysis across Python and data-science domains.
- Contributed to RLHF pipelines for xAI and Meta Llama 4, giving direct experience with how LLMs fail on correctness, instruction following, and reasoning tasks.
- Integrated TTS and STT models into PreCall AI, a production AI calling workflow, strengthening end-to-end model integration experience.
- Translated business requirements into applied AI workflows, drawing on prior customer-facing experience in education technology sales.
Business Development Associate | Byju's2020 - 2022
- Managed high-volume B2C sales of online education products with weekly revenue targets.
- Developed consultative discovery, stakeholder communication, and objection-handling skills across a large customer base.
Selected Projects
rag-evaluation-workbenchhttps://github.com/yashowardhansinghtomar/rag-evaluation-workbench
- Built a RAG evaluator that ingests markdown corpora, chunks documents, retrieves evidence, and scores answers for citation coverage and groundedness.
- Reports retrieval recall@k, citation precision/recall, required-fact coverage, grounded answer rate, and failure tags.
PDF-Question-Answering-Systemhttps://github.com/yashowardhansinghtomar/PDF-Question-Answering-System
- Built a PDF ingestion and document-question-answering workflow using chunking, vector retrieval, and grounded response generation.
real_estate_chatbot_ollamahttps://github.com/yashowardhansinghtomar/real_estate_chatbot_ollama
- Created an offline real-estate assistant with local LLM execution, retrieval-backed FAQ behavior, and multi-modal routing concepts.
Skills
- RAG: Document ingestion, chunking, retrieval, citations, grounded answer evaluation
- LLM Apps: LangChain, OpenAI API, Groq, Ollama, prompt workflows, local-first assistants
- Data / Eval: JSONL eval sets, failure tags, Markdown/JSON reporting, pairwise response review
- Interfaces: Streamlit, Gradio, CLI tools
Education
Prestige Institute of Management & Research
Bachelor of Business Administration (BBA), Global Marketing & Brand Management | 2017 - 2020
Certifications
Data Analyst - Edubridge | Generative AI - Debugshala