Sambhaji Balaji Waghmare
I’m Sambhaji Balaji Waghmare, an AI Data Annotator and AI Evaluator focused on improving the quality of LLM and agent outputs through careful rubric-based review.
About
I’m Sambhaji Balaji Waghmare, an AI Data Annotator and AI Evaluator focused on improving the quality of LLM and agent outputs through careful rubric-based review. I enjoy validating task instructions, checking model responses for correctness and relevance, and hunting for edge cases or failure modes so training and evaluation workflows stay reliable and consistent.
I also bring a software engineering background from my internship experience at TCS, where I supported debugging, testing, and QA using Java, Spring Boot, and SQL. Alongside my AI work on projects like Handshake AI (Project Dynamo) and Aether AI, I collaborate through Git/GitHub workflows, write structured feedback for iteration, and verify behavior using containerized validation setups with Docker/WSL2 and Harbor.
- 1 year's experience
- Joined October 2026
Skills· 10
Languages
- Marathi (Marāṭhī)Fluent
- HindiAdvanced
- EnglishAdvanced
Industry experience
- Software & Internet
- Education
- Professional Services
- Computers & Electronics
Work experience· 6 roles
Supported AI benchmark/task-development workflows involving temporal consensus clustering and evaluation logic. Authored and validated task instructions, solution logic, test cases, and structured input/output expectations for AI-agent evaluation. Used Docker/WSL2 and Harbor-based validation workflows to run and verify task environments and agent outputs. Reviewed implementation behavior for correctness, edge cases, and consistency with task requirements, documenting issues for iteration and maintaining passed criteria during Git/GitHub review cycles (branches, commits, and PR feedback).
Contributed to model-output review for training/evaluation workflows by assessing responses against project-specific task guidelines. Applied consistent criteria for response quality, correctness, relevance, and instruction adherence, and provided structured feedback for improvement. Focused on detailed annotation instructions, with careful attention to errors, inconsistencies, and edge-case failures.
Supported AI benchmark/task-development workflows with temporal consensus clustering and evaluation logic. Authored and validated task instructions, solution logic, test cases, and structured input/output expectations for AI-agent evaluation. Used Docker/WSL2 and Harbor-based validation workflows to run and verify environments and agent outputs. Reviewed implementations for correctness, edge cases, and failure modes, and documented issues for iteration while preserving passed criteria via Git/GitHub PR workflows.
Education· 9
Hire Sambhaji Balaji Waghmare today
Post a job and invite Sambhaji, or contact them directly. No hidden commission.