Rohit A.

Key Skills

Software

Don't disclose

Other

Top Subject Matter

Agentic LLM decision-support for taxi fleet dispatch

time-series forecasting and evaluation

Education/tutoring for middle-grade students (curriculum content creation)

Top Data Types

Text

Top Task Types

Function Calling

Classification

Freelancer Overview

Research Intern — RideCopilot (Agentic LLM Copilot for Taxi Fleet Dispatch). Brings 3+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Don't disclose and Other. Education includes Master of Technology, Indian Institute of Technology Roorkee (2026) and Bachelor of Technology, Indraprastha Institute of Information Technology, Delhi (2024). AI-training focus includes data types such as Computer Code, Programming, and Text and labeling workflows including Function Calling, Computer Programming, and Coding.

Labeling Experience

Research Intern - RideCopilot

TextClassification

You worked as a Research Intern to build an agentic LLM decision-support system for taxi fleet dispatch. You translated multi-stakeholder natural language into bounded DecisionSpec updates consumed by a solver-grounded dispatch optimizer. You performed end-to-end system validation, integrating forecasting and classification metrics with production-minded evaluation practices. • Architected an agentic decision-support workflow for controllable and auditable fleet decisions • Integrated Chronos time-series demand forecasting with spike detection and reported MAE, R², ROC-AUC, and F1 • Validated the approach on the NYC TLC simulator and analyzed order-rejection and earnings fairness outcomes • Authored a first-author paper describing the system and its deployment-oriented results

2025 - 2025

Research Intern — RideCopilot (Agentic LLM Copilot for Taxi Fleet Dispatch)

Don't discloseFunction Calling

Conducted agentic LLM evaluation and tool-using system development for a taxi dispatch copilot, including simulation-based validation. Implemented structured decision updates that serve as labeled outputs for downstream solver consumption. Built and tested components combining forecasting and classification metrics to assess model behavior and decision quality. • Integrated Chronos demand forecasting with spike detection metrics (MAE, R², ROC-AUC, F1). • Validated end-to-end on NYC TLC simulator with order-rejection and earnings distribution metrics. • Produced bounded, auditable DecisionSpec updates consumed by a dispatch optimizer. • Authored a first-author publication describing the agentic, LLM-based conversational system.

2025 - 2025

Teaching Volunteer — Jagrukta Abhiyan

OtherText

Created structured teaching materials and delivered lessons for middle-grade students as part of an AI/ML-oriented learning initiative. Developed study content end-to-end, including curriculum structure and original materials. Focused on instructional guidance and learning outcomes rather than dataset annotation. • Designed and delivered structured lessons. • Owned curriculum content end-to-end. • Produced original study materials for students. • Supported student learning through tutoring sessions.

2023 - 2023

Education

Work History