Internship AI Engineer resume example

• Under supervision, built an evaluation set separating retrieval errors from unsupported answers, checked document permissions before retrieval, and recorded citation accuracy alongside latency and token cost.
• The team built a Python RAG pipeline for a support knowledge base, preserving document permissions and citation links through retrieval and reranking; I supported research preparation, validation, and documentation under mentor review. The team made unsupported answers traceable to missing or stale source documents; confirmed with my mentor that my contribution was limited to the assigned support and validation work.
• The team implemented automated testing with an LLM evaluation set from anonymized support questions, separating answer correctness, citation coverage, and prompt injection resistance; I supported research preparation, validation, and documentation under mentor review. The team blocked model releases that improved fluency but regressed permission handling; confirmed with my mentor that my contribution was limited to the assigned support and validation work.
• Under supervision, owned a RAG evaluation using synthetic clinical questions and Python fixtures, resolved unsupported citations and completed human review before a limited demonstration.
• Under supervision, built automated testing for prompt versions using Docker and CI/CD, reduced repeated citation failures by 34% on the held-out fixture and published observability checks.

• In a mentor-reviewed simulation, built a Python RAG service over permission-filtered support documents. Compared retrieval depth and prompt engineering variants using fixed answer and citation tests; selected the configuration that resolved untraceable citations without exposing restricted documents.
• As a second supervised exercise, added automated testing to CI/CD for refusal behavior, citation validity, and prompt injection; connected monitoring to the same failure categories so a fluent but unsupported answer could block a release.
• Under supervision, the team added monitoring for token usage and p95 latency by request type, then routed simple requests to a smaller model with an explicit fallback; I supported research preparation, validation, and documentation under mentor review
• Generated synthetic source records with duplicates, late arrivals and corrected values; wrote assertions for row counts and key uniqueness, and recorded the expected effect of each case on the reported metric.
• Compared the analytical output with a manually calculated reference table, traced differences to a transformation step, and retained a data dictionary and rerun instructions alongside the corrected query.
• Owned the synthetic-data validation using a manually calculated reference, resolved duplicate-key inflation and completed a notebook that reproduces the corrected totals.

Relevant coursework: Algorithms, operating systems, databases, computer networks

• Published an independent methods note using reproducible queries, explaining the data grain, excluded records and sensitivity of the result to a changed denominator.