App LogoApp name
YM

Yuvraj mehta

Open to work2 years experience

Data Scientist

Banglore, IndiaTarget Roles: Backend Engineer • Full-Stack Developer • Frontend Specialist

Data Scientist specializing in AI Agents, RAG, and LLM fine-tuning.

GitHub

Standing Rank

Rank Not Available

Developer Badges

No badges earned yet

Skills & Technologies

24 skills
pythonsqlgitfastapilangchainllamaindextensorflowstreamlitlanggraphgpt-4sentence-bertwhisperfine tuningopenai apichromadbnltkspacydockerpandasnumpymatplotlibseabornscikit-learnkeras

Work Experience

Data Scientist

Tech Relieve

Nov 2025 - Present Bengaluru, Karnataka
  • Built and deployed a production voice AI agent that autonomously placed 50,000+ outbound calls in a month (13,000+ connected), running real-time qualification conversations over SIP telephony with a dual voice pipeline — Gemini Live (native audio, sub-400ms latency) and a cascade STT→LLM→TTS path — switchable per tenant on a multi-tenant platform
  • Designed the post-call processing pipeline: LLM-based qualification extraction + a weighted 0–100 lead-scoring engine with cumulative cross-call blending, producing 25,000+ structured lead records per month
  • Hardened the agent for real-world telephony — filler-audio latency masking when the LLM stalls>1.5s, and a silence watchdog that nudges an unresponsive caller and ends the call on continued silence to avoid runaway call cost — failure modes surfaced only by running thousands of live calls

Data Scientist

Aagati Pvt. Ltd.

Nov 2024 - Nov 2025 Bengaluru, Karnataka
  • Built a production multi-tenant RAG system for an employee-benefits platform — hybrid retrieval (BM25 + Pinecone), a dual-index strategy, and a streaming voice interface (STT + multi-provider TTS) — answering benefit inquiries at 2–3s latency across partner organizations
  • Developed an AI-powered ATS scoring engine combining a custom-trained spaCy NER (500 annotated resumes) with a fine-tuned Sentence-BERT matcher (6,000 resume–JD pairs) and a weighted scoring algorithm, plus a Twilio/WhatsApp candidate-engagement bot for automated interview scheduling

Projects

SaaS Helpdesk Voice Agent

LangGraphLiveKitFastAPIPostgres
  • Built a multi-agent voice helpdesk demonstrating modern agentic patterns — a config-driven router/triage agent performs mid-call handoff to Account, Billing, and Technical specialists (LiveKit), with tool and agent registries that let agents be defined as data (prompt + tools + handoff targets), requiring no code changes to add one
  • Engineered an async post-call LangGraph workflow — a stateful graph (summarize→classify intent→sentiment →conditional routing→ticket/persist) with a Postgres checkpointer and deterministic fallbacks, keeping the real-time call loop fast while moving agentic reasoning to the async layer

RAG over SEC 10-K Filings

RAGFastAPIPineconeGemini
  • Built an end-to-end RAG system over SEC 10-K filings with hybrid retrieval (BGE vector + BM25 fusion) and cross-encoder reranking, returning streamed, source-cited answers via FastAPI/SSE
  • Engineered domain-aware hierarchical chunking (split by SEC Item headers, table-preserving) on a clean interface/factory dependency-injection architecture (swappable LLM, vector store, and reranker via config); added voice + multilingual querying (Sarvam STT + auto-translate)

Leaderboard Standings

Leaderboard Position Pending

Global test scores, peer standing percentiles, and algorithm leaderboard ranks are updated dynamically.

Assessment Highlights

Assessments Not Completed

Coding evaluations, system assessment results, and conceptual score badges will appear here after taking a test.

AI Collaboration Score

AI Collaboration Score Pending

Developer coding behavior, assistant cooperation, and AI pair-programming indicators are evaluated during live coding sessions.

Role Compatibility Profile

Role Compatibility Analysis Pending

Custom matching reports, candidate role compatibility percentiles, and core engineer strength profiles are processed once conceptual code screenings are complete.

Achievements

Achievements Not Earned

Special rewards, developer badges, system recognition certificates, and conceptual milestones will display here.

About Details

Professional Bio

Data Scientist with expertise in building production-grade AI agents, RAG systems, and fine-tuning LLMs. Experienced in deploying scalable voice automation, computer vision frameworks, and NLP solutions. Passionate about leveraging advanced machine learning to solve complex business challenges.

M.TECH in Industrial Engineering and Management

National Institute of Technology Calicut, Kerala (2022 - 2024)

B.TECH in Industrial and Production Engineering

G.B.P.U.A.T Pantnagar, Uttarakhand (2017 - 2021)

Languages: English, Hindi