Kiran Kumar (Kumar) K.
Applied NLP: Word-Level Encoding for Smarter Event Predictions
Experience
Nov 2024 - May 2025
7 monthsGermany
Applied NLP: Word-Level Encoding for Smarter Event Predictions
University of Siegen
- Engineered 12+ Seq2Seq models (LSTMs/GRUs) to train an AI model, supporting AI Agent Evaluation Analyst and online projects in complex systems
- Conducted 15+ experiments to improve forecasting accuracy by 28%, applying analytical thinking and testing models for QA and edge case coverage
- Researched 20+ papers as a consultant to guide Large Language Model - LLM design, ensuring logical implications and domain of expertise alignment
- Saved 40% compute time via model compression with reusable PyTorch framework, aiding developers and writers to suggest refinements and improve policies
May 2024 - Nov 2024
7 monthsGenerative AI Applications with RAG and LangChain
IBM, Coursera
- Built a Large Language Model - LLM Q&A bot using LangChain, enhancing retrieval efficiency by 35% for online projects in domain of expertise
- Applied prompt engineering to train an AI model, improving response accuracy by 25% and reducing irrelevant outputs by 30%, supporting AI Agent Evaluation Analyst tasks
Nov 2023 - Apr 2024
6 monthsGermany
AI Engineer (Master Thesis)
ABB AG Research Center
- Designed and deployed 32+ self-supervised models to train an AI model using Python and TensorFlow for AI Agent Evaluation Analyst tasks in complex systems
- Built and managed Azure ML data pipelines supporting 3 teams, enabling QA and developers to review evaluation tasks and improve edge case coverage
- Developed pretext tasks for 100K+ samples, annotating cause-effect relationships to enhance Large Language Model - LLM accuracy and reduce manual labeling by 70%
- Delivered strong business outcomes by applying analytical thinking and logical implications to define clear expected behaviors and gold standards in diagnostics
Apr 2023 - Oct 2023
7 monthsGermany
Full Stack Developer
ABB AG Research Center
- Built full-stack tools using Python, JavaScript, React to train an AI model, integrating 10+ datasets/APIs for AI Agent Evaluation Analyst tasks
- Collaborated in 10-person team as consultant, improving workflows, reducing errors by 20%, ensuring QA and edge case coverage in complex systems
- Delivered 10+ online projects under deadlines, reviewing evaluation tasks, identifying inconsistencies, and supporting 4+ monthly stakeholder decisions in domain of expertise
- Developed web app integrating 4+ renewables, annotating cause-effect relationships, enhancing client energy use by 35%, and defining clear expected behaviors for policies
AI-Powered Resume and Cover Letter Tailoring Website
- Developed a web app with LLMs, LangChain, RAG, Pinecone for AI Agent Evaluation Analyst tasks, improving online projects in the domain of expertise
- Built and deployed AI model system with cloud pipelines, enhancing QA, annotating cause-effect relationships, and supporting full-time opportunities
Personal AI Chatbot portfolio
- Built an interactive chatbot portfolio using LLMs, React, Flask, Tailwind, RAG, chromaDB, enhancing AI Agent Evaluation Analyst capabilities and recruiter engagement by 60%
- Integrated Git, Docker, CI/CD pipelines for QA, enabling train an AI model, improving edge case coverage and cutting recruiter review time by 40%
Summary
AI Agent Evaluation Analyst with expertise in training Large Language Models (LLMs), reviewing evaluation tasks for logic, completeness, and realism, identifying inconsistencies, defining gold standards, annotating reasoning paths, and applying analytical thinking to complex systems and policies.
- Analytical Thinking: Engineered 12+ Seq2Seq models, influencing 3+ teams and 2 theses, boosting forecasting accuracy by 28%.
- Attention to Detail: Reviewed evaluation tasks and scenarios for LLM logic, realism, and completeness, identifying inconsistencies and missing assumptions.
- Familiarity with Structured Data Formats: Built and deployed AI resume systems with cloud pipelines and prompt design, cutting tailoring time by 50%.
Languages
Hindi
NativeTelugu
NativeEnglish
AdvancedGerman
IntermediateEducation
Oct 2023 - May 2025
University of Siegen
Master of Science, Specialization in AI · Mechatronics · Germany
MLR Institute of Technology, JNTUH
Bachelor of Technology · Mechanical Engineering · India
Certifications & licenses
The Systems Modeling Language (SysML®) v1.6
Udemy
AWS Cloud Technical Essentials
AWS & Coursera
Generative AI Engineering With LLMs Specialization
IBM & Coursera
Building Deep Learning Models With TensorFlow
IBM & Coursera
Deep Neural Networks With PyTorch
IBM & Coursera
Introduction To Machine Learning By Andrew Ng
Stanford Online & Coursera
IBM Certified Generative AI Specialist
IBM
Need a freelancer? Find your match in seconds.
Try FRATCH GPT More actions
Similar Freelancers
Discover other experts with similar qualifications and experience
- English
- Deutsch
2025 © FRATCH.IO GmbH. All rights reserved.