For employers

Hire this AI Trainer

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

LoginCreate account
Invite to Job
Suman Gowda

Suman Gowda

Certified in Data Analyst | Data scientist| Artificial Engineering| AI Trainer |Certified in Data Science by State University of New York at Potsdam |

INDIA flag
Bangalore, India
$10.00/hrIntermediateRoboflowAws SagemakerLabelbox

Key Skills

Software

RoboflowRoboflow
AWS SageMakerAWS SageMaker
LabelboxLabelbox
Other

Top Subject Matter

Large Language Models
Conversational AI
LLM Evaluation

Top Data Types

TextText
DocumentDocument
ImageImage

Top Task Types

Fine Tuning
Prompt Response Writing SFT
Object Detection
Classification

Freelancer Overview

Supervised Fine-Tuning Data Labeler for Llama-2-7B-Chat. Brings 5+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Other. Education includes Nanodegree Certification Program, The State University of New York at Potsdam (2023) and Professional Training Course, 360DigiTMG (2023). AI-training focus includes data types such as Text, Document, and Image and labeling workflows including Fine-tuning, Prompt + Response Writing (SFT), and Object Detection.

IntermediateKannadaHindiEnglish

Labeling Experience

Prompt Engineer and Data Annotator for Gemini Pro ATS

OtherDocumentPrompt Response Writing SFT
Developed and labeled data for an Application Tracking System using Google’s Gemini Pro LLM, focusing on resume evaluation tasks. Created, annotated, and labeled prompt-response pairs for ATS scenarios and feedback generation. Designed evaluation rubrics and sample outputs to guide automated scoring systems. • Crafted sample resumes and job descriptions for LLM evaluation. • Annotated outputs with scoring and qualitative feedback labels. • Engineered prompts to simulate realistic HR screening tasks. • Evaluated LLM outputs for accuracy and relevance to job alignments.

Developed and labeled data for an Application Tracking System using Google’s Gemini Pro LLM, focusing on resume evaluation tasks. Created, annotated, and labeled prompt-response pairs for ATS scenarios and feedback generation. Designed evaluation rubrics and sample outputs to guide automated scoring systems. • Crafted sample resumes and job descriptions for LLM evaluation. • Annotated outputs with scoring and qualitative feedback labels. • Engineered prompts to simulate realistic HR screening tasks. • Evaluated LLM outputs for accuracy and relevance to job alignments.

2024 - 2024

Supervised Fine-Tuning Data Labeler for Llama-2-7B-Chat

OtherTextFine Tuning
Conducted supervised fine-tuning on the Llama-2-7B-Chat model to replicate corporate boss communication for an enterprise client. Iteratively adapted responses and optimized prompt engineering to enhance the AI’s persona emulation. Used GradientAI to run training, leveraging both model adaptation and evaluation processes. • Labeled and curated dialogue data to align with targeted corporate communication styles. • Designed prompts and evaluated output responses for desired behavioral accuracy. • Applied supervised methods to continuously improve chat quality and consistency. • Assessed model responses and provided annotations for error correction and further training.

Conducted supervised fine-tuning on the Llama-2-7B-Chat model to replicate corporate boss communication for an enterprise client. Iteratively adapted responses and optimized prompt engineering to enhance the AI’s persona emulation. Used GradientAI to run training, leveraging both model adaptation and evaluation processes. • Labeled and curated dialogue data to align with targeted corporate communication styles. • Designed prompts and evaluated output responses for desired behavioral accuracy. • Applied supervised methods to continuously improve chat quality and consistency. • Assessed model responses and provided annotations for error correction and further training.

2023 - 2024

Image Classification Data Annotator for Dietary Analysis AI

OtherImageClassification
Developed an advanced nutrition analysis tool using Gemini Pro Vision and Python for automated image-based dietary evaluation. Labeled and annotated food images for caloric and nutrient computation tasks, validating AI outputs against expert dietary standards. Ensured dataset quality and diversity to enhance recognition and prediction performance. • Curated image datasets of various food items for classification. • Labeled nutrient type, food group, and portion sizes for training purposes. • Implemented annotation guidelines for consistent dataset labeling. • Conducted quality assurance and reviewed AI predictions on labeled sets.

Developed an advanced nutrition analysis tool using Gemini Pro Vision and Python for automated image-based dietary evaluation. Labeled and annotated food images for caloric and nutrient computation tasks, validating AI outputs against expert dietary standards. Ensured dataset quality and diversity to enhance recognition and prediction performance. • Curated image datasets of various food items for classification. • Labeled nutrient type, food group, and portion sizes for training purposes. • Implemented annotation guidelines for consistent dataset labeling. • Conducted quality assurance and reviewed AI predictions on labeled sets.

2023 - 2023

Data Labeler for Drowsiness Detection and Object Recognition (YOLOv5)

OtherImageObject Detection
Engineered and labeled custom datasets for YOLOv5-based drowsiness detection and object recognition system. Applied bounding box labeling to annotate states of alertness, drowsiness, and road objects in video frames. Fine-tuned the YOLOv5 model on the labeled image data to enhance system safety and accuracy. • Collected and pre-processed images from road safety datasets. • Annotated bounding boxes on faces and road objects using computer vision tooling. • Evaluated model predictions and iterated label corrections for improved detection. • Integrated annotated data into fine-tuning pipelines for continuous improvement.

Engineered and labeled custom datasets for YOLOv5-based drowsiness detection and object recognition system. Applied bounding box labeling to annotate states of alertness, drowsiness, and road objects in video frames. Fine-tuned the YOLOv5 model on the labeled image data to enhance system safety and accuracy. • Collected and pre-processed images from road safety datasets. • Annotated bounding boxes on faces and road objects using computer vision tooling. • Evaluated model predictions and iterated label corrections for improved detection. • Integrated annotated data into fine-tuning pipelines for continuous improvement.

2023 - 2023

Education

3

360DigiTMG

Professional Training Course, Data Science and Artificial Intelligence

Professional Training Course
2023 - 2023
T

The State University of New York at Potsdam

Nanodegree Certification Program, Data Science and Machine Learning

Nanodegree Certification Program
2023 - 2023

Work History

I

Innodatatics

Artificial Intelligence Engineer

Bangalore
2023 - Present
A

Aspireit

Freelance AI Engineer

N/A
2022 - 2023