AI Research Engineer (Early Career)

Executive Summary

The candidate is a self-taught, entrepreneurial software engineer with genuine production delivery experience, but they are not a strong match for this AI Research Engineer role as defined. Their AI work is largely surface-level API integration and workflow automation rather than the deep LLM systems, RAG pipelines, vector search, and agentic orchestration the role centers on. They also lacks the formal CS/Math university background the role specifies, having completed a full-stack coding bootcamp instead. Compounding these technical gaps, there are notable inconsistencies between their LinkedIn and resume employment history that reduce confidence in the application's accuracy. While the candidate shows promise as a software engineer and could potentially develop ML skills over time, they are not ready for this specific role today. The recommendation is NOT_FIT, though a future application after dedicated upskilling in LLM engineering (LangGraph, RAG, vector databases) could look quite different.

Assessment Reasoning

The candidate does not meet the minimum bar for this AI Research Engineer role for three compounding reasons. First, the technical skills gap is substantial: the role explicitly requires hands-on experience with RAG pipelines, vector databases, LangGraph, AI agents, and LLM fundamentals (embeddings, fine-tuning, evaluation) — none of which appear in the candidate's resume or project portfolio in any meaningful way. Their 'AI' experience is OpenAI API calls, n8n automation, and basic Scikit-learn forecasting, which fall well short of the applied ML engineering depth required. Second, the education requirement (strong CS, Math, or ML university program) is not met — Moringa School is a reputable full-stack bootcamp but is not a CS/Math degree program and does not provide the ML theoretical grounding the role expects. Third, the LinkedIn/resume date discrepancies across multiple roles (NEOTRILLI, Simplyinfo) introduce a credibility concern that would need to be resolved before any offer could be considered. The candidate scores approximately 42/100 overall, falling in the NOT_FIT range (below 50). Their software engineering fundamentals and production delivery experience are genuine assets, but they do not compensate for the core technical skill misalignment with this specific role.

Interview Focus Areas

No code sample was submitted and no GitHub link was verified, making a direct code quality assessment impossible. Based on project descriptions alone, the candidate demonstrates competent full-stack development but no evidence of production-quality ML/LLM engineering code. The absence of submitted code is itself a mild red flag for a role emphasizing production engineering over notebooks.

The candidate is a capable full-stack engineer and data analyst with 4 years of practical experience, but their AI/ML exposure is shallow relative to what this role demands. Their work with 'AI' is largely API integration and workflow automation, not the RAG pipelines, agentic systems, LangGraph orchestration, or vector database engineering central to this position. The education requirement (strong CS/Math university program) is also unmet, as their background is a full-stack bootcamp.