Enterprise AI teams moving from proof-of-concept to production frequently search to hire a freelance LLM evaluation, Ragas, DeepEval, Arize Phoenix & LangSmith programmer to measure and eliminate hallucinations, tool-call failures, and prompt regressions. I am Aldian Fazrihady, a Senior Freelance AI Application Engineer and Full Stack Programmer available via Upwork.
Core LLM Evaluation & AI Observability Capabilities
- Automated Golden Dataset & CI/CD Evals: Building deterministic and LLM-as-a-Judge evaluation pipelines with Ragas, DeepEval, Promptfoo, and LangSmith to score faithfulness, context recall, answer relevance, and tool-selection accuracy on every pull request.
- OpenTelemetry & Production Tracing: Instrumenting LangGraph, Google ADK, and OpenAI/Anthropic SDK workflows with Arize Phoenix, Langfuse, and LangSmith to trace token latency, cost per tenant, and retrieval quality in real time.
- Red-Teaming & Prompt Injection Guardrails: Auditing AI agents against adversarial jailbreaks, indirect prompt injections, and PII leakage.
Frequently Asked Questions
How do we stop prompt or model changes from breaking existing AI features?
I curate a 100+ scenario golden evaluation dataset from real production edge cases and wire automated regression gates into GitHub Actions so any drop in faithfulness or JSON schema compliance blocks deployment.
How can we hire you as a freelance LLM evaluation programmer?
Message me on Upwork.