Data & Context Engineer Attachee

About the role

Data & Context Engineering is the discipline of curating, structuring and governing the knowledge that AI models and agents use: retrieval pipelines, context windows, knowledge bases and their data quality standards. As a Data & Context Engineer Attachee at Code Particles, you will build the context layer for the company's agent fleet, keeping knowledge fresh and correct and measuring how context quality affects agent performance. The internship is based in Nairobi.

Key responsibilities

• Design and build retrieval pipelines (RAG) that feed agents accurate, current knowledge
• Curate and structure knowledge bases: chunking, metadata, embeddings and freshness policies
• Engineer context windows: what an agent sees, in what order and within what budget
• Measure how context quality affects agent accuracy and run experiments to improve it
• Build data quality checks and lineage tracking for the sources agents depend on
• Detect and repair knowledge drift: stale documents, contradictions and coverage gaps
• Apply data governance to AI context: what is allowed in, PII handling and access boundaries
• Document context architecture decisions and present findings to the engineering team

Required profile

• Currently pursuing a degree in Computer Science, Data Science, Information Science, Statistics or a related field, requiring industrial attachment for academic credit
• Institution must provide an attachment letter and an evaluation framework (logbook, supervisor assessment or equivalent)
• Experience consuming APIs and working with structured and unstructured data
• Clear written communication

Required skills

• Proficiency in Python and comfort with SQL
• Understanding of how LLMs consume context (tokens, context windows, embeddings)
• Working knowledge of Git and version control workflows
• Nice to have: vector databases (pgvector, Pinecone, Weaviate) or search engines, dbt, Airflow or cron scripts, Kenya DPA and GDPR, documentation and wiki experience