For employers

Hire this AI Trainer

Sign in or create an account to invite AI Trainers to your job.

LoginCreate account
Invite to Job
A

Anjali Laishram

Software Engineer | AI Training (RLHF)

India flagBangalore, India
$50.00/hrExpertOtherLabelbox

Key Skills

Software

Other
LabelboxLabelbox

Top Subject Matter

AI Coding Model Training
AI Conversational Agent Training

Top Data Types

TextText
VideoVideo
Computer Code ProgrammingComputer Code Programming

Top Task Types

RLHF
Prompt Response Writing SFT
Evaluation Rating
Transcription
Computer Programming Coding
Function Calling

Freelancer Overview

Software Engineer | AI Training (RLHF). Brings 4+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Other, Internal, and Proprietary Tooling. Education includes Bachelor of Computer Applications, SRN Adarsh College (2017). AI-training focus includes data types such as Computer Code, Programming, and Text and labeling workflows including RLHF and Prompt + Response Writing (SFT).

ExpertEnglish

Labeling Experience

AI Code Evaluation and Model Training

Computer Code ProgrammingComputer Programming Coding
AI code evaluation project on Anthropic's feedback platform. I evaluate two AI models side by side on real open source GitHub repositories. Tasks include writing coding prompts, reviewing model generated code turn by turn, rating responses across 7 quality dimensions (logic, correctness, naming, organization, interface design, error handling, documentation, production readiness), writing detailed technical feedback, and selecting the better model with justification. I also review model attempts at fixing real GitHub issues, checking for correctness, regressions and code quality. The project involves working with Python, JavaScript and TypeScript codebases of varying size and complexity. Quality measures include structured evaluation rubrics, actionable improvement suggestions with specific file and function references, and multi turn conversations guiding models toward merge ready code.

AI code evaluation project on Anthropic's feedback platform. I evaluate two AI models side by side on real open source GitHub repositories. Tasks include writing coding prompts, reviewing model generated code turn by turn, rating responses across 7 quality dimensions (logic, correctness, naming, organization, interface design, error handling, documentation, production readiness), writing detailed technical feedback, and selecting the better model with justification. I also review model attempts at fixing real GitHub issues, checking for correctness, regressions and code quality. The project involves working with Python, JavaScript and TypeScript codebases of varying size and complexity. Quality measures include structured evaluation rubrics, actionable improvement suggestions with specific file and function references, and multi turn conversations guiding models toward merge ready code.

2026 - Present

Software Engineer | AI Training (RLHF)

OtherRLHF
This role involved working on Reinforcement Learning from Human Feedback (RLHF) to improve AI model performance by generating and evaluating coding-related training data. I designed prompts and performed side-by-side comparisons of AI model outputs for accuracy, focusing on high-quality data generation. The work utilized multiple programming languages and contributed to large language model advancements. • Generated and evaluated coding-based training datasets for large language models. • Designed prompts to assess and refine model outputs for correctness and quality. • Used Python, JavaScript, TypeScript, Java, and SQL to create realistic code samples and evaluations. • Facilitated the improvement of AI accuracy and quality in code generation applications.

This role involved working on Reinforcement Learning from Human Feedback (RLHF) to improve AI model performance by generating and evaluating coding-related training data. I designed prompts and performed side-by-side comparisons of AI model outputs for accuracy, focusing on high-quality data generation. The work utilized multiple programming languages and contributed to large language model advancements. • Generated and evaluated coding-based training datasets for large language models. • Designed prompts to assess and refine model outputs for correctness and quality. • Used Python, JavaScript, TypeScript, Java, and SQL to create realistic code samples and evaluations. • Facilitated the improvement of AI accuracy and quality in code generation applications.

2024 - Present

Agentic LLM Trainer

TextPrompt Response Writing SFT
As an Agentic LLM Trainer, I created realistic user-agent conversations based on complex prompts and instructions to train chat agents. The labeling process included prompt engineering, conversation generation, evaluation, feedback, and iterative improvements. I ensured that system and user message prompts captured varied tool integrations and contextual information. • Generated multi-turn chat data reflecting agent interactions with various tools. • Engineered prompts and synthesized varied user goals and behaviors. • Performed evaluation, correction, and rating of model-generated outputs. • Used proprietary tools and GPT-4 to simulate conversation and feedback cycles.

As an Agentic LLM Trainer, I created realistic user-agent conversations based on complex prompts and instructions to train chat agents. The labeling process included prompt engineering, conversation generation, evaluation, feedback, and iterative improvements. I ensured that system and user message prompts captured varied tool integrations and contextual information. • Generated multi-turn chat data reflecting agent interactions with various tools. • Engineered prompts and synthesized varied user goals and behaviors. • Performed evaluation, correction, and rating of model-generated outputs. • Used proprietary tools and GPT-4 to simulate conversation and feedback cycles.

2025 - 2025

Education

S

SRN Adarsh College

Bachelor of Computer Applications, Computer Applications

Bachelor of Computer Applications
2013 - 2017

Work History

N

Nexus Software

Software Engineer

Bangalore
2022 - 2025