AI Engineer

About the Role

This entry-level AI Engineer will apply developing technical knowledge in Software Engineering and Generative AI. Responsibilities include analytic, testing, modeling, and development tasks. The role will help design, build, and deploy AI-enabled software solutions, including LLM-powered applications, retrieval-augmented generation (RAG) pipelines, and model-serving APIs.

Key Responsibilities

• Perform assigned technical and analytic tasks using established methods and tools.
• Document methods and results, and escalate novel or ambiguous issues.
• Support development of AI-enabled applications and fix straightforward defects.
• Build and maintain components of LLM-based solutions, including API integrations, prompt and tool/function-calling workflows, and basic RAG pipelines.
• Clean, transform, and analyze data using Python, SQL, and data libraries.
• Build simple ETL steps and visualizations.
• Execute assigned test cases, including unit tests and basic retrieval or model evaluations.
• Reproduce reported issues.
• Serve models and AI services through REST APIs and containers.
• Prepare simple configuration changes and follow deployment runbooks.
• Track experiments and model versions.
• Support monitoring and CI/CD processes.
• Document work and escalation steps in tickets and knowledge base entries.
• Learn codebase, development workflows, and platform fundamentals through pairing and code reviews.
• Produce reliable technical outputs and follow quality and safety standards (including AI guardrails).
• Document reproducible methods and records.
• Communicate progress and risks to project leads.

Requirements & Tech Stack

• Software Engineering
• Generative AI
• Python
• SQL
• NumPy
• pandas
• Matplotlib
• Seaborn
• Plotly
• Pytest
• Git/GitHub
• Jupyter
• VS Code
• Docker basics
• SQL (essential)
• Relational databases: PostgreSQL, MySQL, or SQLite
• SQLAlchemy
• psycopg2
• pandas.read_sql
• NoSQL: MongoDB, Redis
• Vector databases: pgvector, Chroma, FAISS, Pinecone
• ETL basics
• Airflow concepts
• LLM APIs: Anthropic, OpenAI
• LangChain or LlamaIndex (basics)
• MLOps
• Model serving
• FastAPI
• Experiment tracking
• MLflow
• Weights & Biases
• Cloud basics: AWS, Azure, or GCP
• CI/CD basics
• Monitoring
• Model versioning

Qualifications

• Bachelor’s Degree in Computer Science, Data Science, Engineering, Mathematics, or relevant STEM certification.
• Problem decomposition.
• Clear communication.
• Reading documentation.
• Collaborating through code review.
• Explaining results to non-technical people.
• Python and Software Fundamentals: data types, control flow, functions, OOP, comprehensions, generators, decorators, exceptions, file I/O, virtual environments (venv, pip / uv).
• Data libraries: NumPy, pandas (cleaning, joins, groupby, time series).
• Visualization: Matplotlib, Seaborn, Plotly.
• Code quality: type hints, unit testing (pytest), logging, PEP 8, packaging.
• APIs and web: requests, REST concepts, FastAPI or Flask to serve models, async basics.
• Tooling: Git/GitHub, Jupyter, VS Code, Docker basics.
• Databases: SQL (essential) including SELECT, WHERE, JOINs, GROUP BY, HAVING, subqueries, CTEs, window functions, indexes, query optimization. Schema design, normalization, primary and foreign keys, transactions.
• Python connectivity: SQLAlchemy, psycopg2, pandas.read_sql.
• NoSQL: MongoDB (documents), Redis (caching).
• Vector databases (AI-specific): embeddings, similarity search, and tools such as pgvector, Chroma, FAISS, Pinecone.
• Data pipelines: ETL basics, data warehouses, Airflow concepts.
• Modern GenAI Skills: LLM APIs (Anthropic, OpenAI), structured outputs, tool/function calling. RAG: chunking, embeddings, vector search, retrieval evaluation. Frameworks: LangChain or LlamaIndex (know the basics, but understand what's underneath). Agents: tool use, multi-step workflows, basic evaluation, and guardrails.
• MLOps and Deployment: Model serving (FastAPI, Docker). Experiment tracking (MLflow, Weights & Biases). Cloud basics (AWS, Azure, or GCP). CI/CD basics, monitoring, model versioning.