Siva Rajesh V.Private profile: Upwork shows its earnings only to signed-in members, so we check it every few weeks, not daily. MRR here is a rough estimate (Upwork's six-month earnings ÷ months active). Last read Nov 5, 2026.

I help healthcare, pharma and medical-device teams validate computerized systems — and the AI tools now entering them — so they are ready for FDA and EU inspection. 14 years at GE HealthCare in regulated engineering: Computer System Validation (CSV), FDA 21 CFR Part 11, GxP and GAMP 5 risk-based validation, with zero FDA audit findings on production systems I supported. Since 2024 I have also built and tested AI systems (RAG, document AI, LLM evaluation), so I can validate AI the way an inspector will look at it. CSV & compliance: • Validation plans, risk assessments and traceability (URS/FS to test evidence) • IQ/OQ/PQ protocols, test scripts and validation summary reports • 21 CFR Part 11 / EU Annex 11 gap assessments: audit trails, e-signatures, access controls, data integrity • Clinical trial systems: EDC/ePRO platform testing and validation (21 CFR Part 11, EU Annex 11, ICH-GCP) • SOPs and audit/inspection readiness • Risk-based validation aligned with FDA CSA and GAMP 5 (2nd Ed.) AI validation & compliant AI builds: • Validation readiness for AI/LLM tools used in GxP or HIPAA contexts: intended use, risk classification, performance testing, drift monitoring, human-in-the-loop controls (aligned with the FDA/EMA AI principles and draft EU GMP Annex 22) • LLM evaluation suites: hallucination testing, regression testing, answer-quality scoring • RAG and document-AI tools with citations, audit trails and access controls for regulated documents • Fine-tuned open-source LLMs (QLoRA/LoRA) for domain accuracy Selected results: • Zero FDA audit findings on production systems I supported • Cut a medical-device validation execution cycle from 21 days to 2 days (90% effort reduction) by automating protocol execution and evidence capture with AI-OCR • Automated 93% of validation test protocols on medical-device platforms, with traceable, audit-ready evidence • Raised a production AI search assistant to a 97% pass rate on a 30-query evaluation suite (5-star Upwork review) • Improved compliance Q&A accuracy from 2.93 to 3.64/5 on held-out evals with a fine-tuned LLM Technology: Python, FastAPI, LangChain, LangGraph, LlamaIndex, OpenAI API, Claude API, pgvector, Pinecone, SQL. Send me your system, its intended use and your inspection timeline. I will reply with a validation approach, deliverables and risks.