Riteek Kumar

Full-stack developer and AI Analyst currently working with Turing on prompt-based debugging and evaluation of Gemini model outputs, with hands-on practice assessing 200+ LLM-generated responses across accuracy, safety, coherence, and instruction-following, using structured rubrics aligned with RLHF principles. Experienced in prompt engineering (zero-shot, few-shot, chain-of-thought) across GPT, Gemini, and Claude, and in building and shipping full-stack applications with React, TypeScript, and Django. Solid foundation in data structures, algorithms, and system design (100+ problems solved).

Seeking an AI Trainer, LLM Evaluator, or Full Stack Developer role to bring careful, detail-oriented evaluation methodology and strong engineering fundamentals to real-world AI systems.

AI Analyst (Part-time / Freelance) at Turing (2026-04 – Present)

• Perform prompt-based debugging of Gemini model outputs, writing and refining prompts to test model responses for accuracy and correctness
• Identify and document model-matching issues, inconsistencies, and errors in structured debugging log files
• Upload debugging logs and evaluation files to a shared drive for review by the QA and reviewer team
• Collaborate on an iterative model-improvement workflow by providing detailed, reproducible error reports

AI & LLM Evaluation Practice at Independent Study & Applied Practice (2026-01 – Present)

• Completed a structured, self-directed program in advanced prompt engineering: zero-shot, few-shot, chain-of-thought, role-based, and constraint-based prompting
• Evaluated 200+ AI-generated responses (ChatGPT, Gemini, Claude) across 8 dimensions — accuracy, relevance, safety, tone, instruction-following, creativity, hallucination detection, and bias
• Developed personal evaluation rubrics for code-quality assessment aligned with RLHF (Reinforcement Learning from Human Feedback) principles
• Studied AI safety fundamentals: harmful-output detection, jailbreak resistance, fairness, and bias mitigation
• Documented 50+ detailed evaluation case studies with structured feedback, simulating production AI-training workflows

C++ & Data Structures Engineering Training at Cipher Schools (2024-05 – 2024-06)

6-week intensive program

• Mastered advanced DSA concepts and competitive-programming techniques
• Built a Hospital Management System applying Binary Search Trees (BST) with O(log n) search/insert operations
• Solved 50+ algorithm problems focused on optimization and real-world applications

Placement Coordinator & Student Leader at Lovely Professional University (2024-01 – 2024-03)

• Coordinated 10+ recruitment drives involving 50+ companies and 200+ student candidates
• Managed candidate profiles, interview scheduling, and placement records using Excel and internal systems
• Demonstrated leadership, time management, and organizational skill under high-pressure deadlines

Bachelor of Technology in Computer Science & Engineering – Lovely Professional University (2022-08 – 2026-05)

Intermediate (CBSE) – Central Board of Secondary Education (2020 – 2022)