For employers

Hire this AI Trainer

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

Invite to Job
Nukamnduwoh Sama

Nukamnduwoh Sama

Software engineer with track record of reworking AI generated code and UIs.

Cameroon flagLimbe, Cameroon
$25.00/hrEntry LevelOther

Key Skills

Software

Other

Top Subject Matter

No subject matter listed

Top Data Types

Computer Code ProgrammingComputer Code Programming
TextText
VideoVideo

Top Task Types

Bounding Box
Computer Programming Coding
Prompt Response Writing SFT
RLHF

Freelancer Overview

With a background in software engineering, I have contributed in projects which required improving prompts for generating code, rewriting code generated by AI and reviewing/fixing code generated by other contributors.

Entry LevelFrenchEnglish

Labeling Experience

Multi-modal prompts

OtherComputer Code ProgrammingText GenerationRLHF
It involved creating multi-modal prompts to train a vision-language model that can generate or edit UI code based on screenshots, and natural language instructions. The goal is to simplify the development process by enabling the model to understand visual inputs and guidance in plain language to produce clean, functional UI code. What was expected: Reasoning (Text): The response should explain the process clearly and logically . The reasoning should be aligned with the prompt and image(s). Generated Code: Check whether the code is written in the correct language. UI design: Did the code build?. Verify that the resulting UI matches the provided image(s) and prompt requirements. Look for any mismatches, broken components, or missing elements. Score was a scale of 1-5 with three review levels. The task could only be moved forward if at a score was 4 for the first level reviewer and if 5 for the next two levels.

It involved creating multi-modal prompts to train a vision-language model that can generate or edit UI code based on screenshots, and natural language instructions. The goal is to simplify the development process by enabling the model to understand visual inputs and guidance in plain language to produce clean, functional UI code. What was expected: Reasoning (Text): The response should explain the process clearly and logically . The reasoning should be aligned with the prompt and image(s). Generated Code: Check whether the code is written in the correct language. UI design: Did the code build?. Verify that the resulting UI matches the provided image(s) and prompt requirements. Look for any mismatches, broken components, or missing elements. Score was a scale of 1-5 with three review levels. The task could only be moved forward if at a score was 4 for the first level reviewer and if 5 for the next two levels.

2024

Coding UI Generation

OtherComputer Code ProgrammingText GenerationEvaluation Rating
The goal of this project is to create prompts and responses that produce visually appealing and fully functioning front-end UI for web development, based on Nextjs and TypeScript. The project required creating a prompt asking the AI to create an application of a solution which we wanted to create. One the application is generated, we were to copy it into our IDEs, made sure it ran, and fix whatever was broken, as well as enhancing the interface for beauty. Below is the scoring rubric used: Score 5 – Excellent (Perfect Annotation) Score 4 – Strong (Passing with Minor Issues) Score 3 – Borderline (Barely passing. Needs revision to meet full quality expectations. Score 1–2 – Failing (Needs Rewrite)

The goal of this project is to create prompts and responses that produce visually appealing and fully functioning front-end UI for web development, based on Nextjs and TypeScript. The project required creating a prompt asking the AI to create an application of a solution which we wanted to create. One the application is generated, we were to copy it into our IDEs, made sure it ran, and fix whatever was broken, as well as enhancing the interface for beauty. Below is the scoring rubric used: Score 5 – Excellent (Perfect Annotation) Score 4 – Strong (Passing with Minor Issues) Score 3 – Borderline (Barely passing. Needs revision to meet full quality expectations. Score 1–2 – Failing (Needs Rewrite)

2024 - 2024

Education

Q

Quantic School Of Business And Technology

Master of Business Administration, Business Administration

Master of Business Administration
2023 - 2024
U

University of Buea

Master of Engineering, Computer Engineering

Master of Engineering
2015 - 2019

Work History

E

Edward

Co-Founder and Technical Lead

Remote
2024 - Present
O

Optimiz

Lead Software Engineer

Remote
2020 - 2024