Yashowardhan Singh Tomar
Voice AI Engineer | Conversational AI | STT/TTS Integration
Indore, India | Open to remote and relocation | yashtomar10122@gmail.com | +91-97549-50809
linkedin.com/in/yashowardhansinghtomar | github.com/yashowardhansinghtomar
Profile
AI engineer focused on voice AI, conversational systems, and LLM-backed assistant workflows. Integrated TTS and STT models into PreCall AI, a production AI-powered calling workflow, and built a public voice-agent framework demo with provider adapters, streaming-style output, latency traces, batch scenarios, and evaluation reports. RLHF experience with xAI and Meta Llama 4 adds model-quality judgment for conversational behavior.
Experience
Jr. AI / Data Science Engineer | Engineer MasterSep 2024 - Jan 2025
- Integrated Text-to-Speech and Speech-to-Text models into PreCall AI, the company's AI-powered calling workflow.
- Worked across the practical gap between demo voice pipelines and production calling behavior, including provider integration and conversational flow concerns.
- Contributed to RLHF workflows for xAI and Meta Llama 4, evaluating model outputs for correctness, instruction following, clarity, and multi-turn behavior.
- Designed prompts and reviewed side-by-side responses, building judgment for conversational quality, failure modes, and response safety.
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
voice-ai-agent-demohttps://github.com/yashowardhansinghtomar/voice-ai-agent-demo
- Built a local voice-agent framework with replaceable STT, assistant, and TTS boundaries.
- Added streaming-style response artifacts, per-stage latency traces, batch scenario evaluation, safety notes, and Markdown/JSON reports.
llm-evaluation-labhttps://github.com/yashowardhansinghtomar/llm-evaluation-lab
- Built a reusable evaluation workbench for scoring LLM responses, useful for conversational assistant quality review.
rag-evaluation-workbenchhttps://github.com/yashowardhansinghtomar/rag-evaluation-workbench
- Evaluates whether assistant answers are grounded in retrieved evidence and correctly cited.
Skills
- Voice AI: STT/TTS integration, provider boundaries, conversational routing, latency tracing
- LLM Quality: RLHF, response evaluation, prompt design, multi-turn behavior review
- Tools: Python, OpenAI API, Groq, LangChain, Hugging Face, Ollama
- Interfaces: CLI tools, Streamlit, Gradio, Slack API
Education
Prestige Institute of Management & Research
Bachelor of Business Administration (BBA), Global Marketing & Brand Management | 2017 - 2020
Certifications
Data Analyst - Edubridge | Generative AI - Debugshala