Utsav D.

Key Skills

Software

Other

Top Subject Matter

AI equity research and financial document grounding (SEC filings, financial APIs, live web)

LLM evaluation

factual verification

Top Data Types

Document

Text

Top Task Types

Evaluation/Rating

Fine-tuning

Prompt + Response Writing (SFT)

RLHF

Freelancer Overview

AI Product Engineer — Midas AI (multi-agent agentic research platform with LLM evaluation/grounding). Brings 7+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include LangGraph and Other. Education includes Bachelor of Engineering in Civil Engineering, Birla Institute of Technology & Science, Pilani (2020). AI-training focus includes data types such as Document and labeling workflows including Evaluation, Rating, and RLHF.

Labeling Experience

AI Product Engineer — Midas AI (multi-agent agentic research platform with LLM evaluation/grounding)

Document

Built and evaluated an LLM-driven multi-agent research pipeline for financial equity research using LangGraph and hybrid RAG sources. Implemented heuristic + LLM-as-judge scoring to verify report claims against ground-truth evidence and enforce citation-level provenance. Designed quality gating across retrieval, planner/coordinator, research, and writer stages to improve factuality and reduce hallucinations. • Scored reports on 8 weighted dimensions using LLM-as-judge • Grounded outputs by checking each claim against filings, financial APIs, and live web evidence • Implemented per-claim citation provenance with deterministic computation and source-priority fusion • Added anti-hallucination layers and reproducible evaluation for citation-level accuracy

2024 - Present

Education

Work History