Role overview
- Implement features in production AI applications, including LLM integrations, prompt workflows, retrieval pipelines, and supporting backend services.
- Develop and maintain components of RAG systems, including data ingestion, chunking, embedding generation, and retrieval logic.
- Write clean, tested, well-documented Python code that meets team standards for quality and maintainability.
- Build internal tools, scripts, and prototypes that accelerate the team's ability to experiment and iterate.
- Run experiments to evaluate model performance, prompt variations, retrieval strategies, and end-to-end system behavior.
- Develop and maintain evaluation datasets, test cases, and regression checks for AI features.
- Analyze production logs and metrics to identify quality issues, latency bottlenecks, and cost optimization opportunities.
- Contribute to incident response and root-cause analysis for AI system issues.
- Stay current with the AI ecosystem by following research, exploring new tools, and bringing useful ideas back to the team.
- Participate actively in code reviews, design discussions, and team rituals, asking questions and offering perspectives.
- Document your work clearly so that teammates can build on it and learn from it.
- Pair with senior engineers on complex problems and gradually take on larger scope as you grow.
- Work closely with product managers, designers, and other engineers to understand requirements and ship features that solve real user problems.
- Communicate progress, blockers, and trade-offs clearly in standups, written updates, and design documents.
- Support other teams by answering questions about AI capabilities and limitations.
Basic qualifications
- 1 to 2 years of professional software engineering experience (internships, co-ops, and substantial open source contributions count).
- Strong programming skills in Python, with familiarity in writing modular, testable code.
- Working knowledge of how large language models behave in practice, including experience calling LLM APIs (OpenAI, Anthropic, Google, or open-weight models) in at least one project.
- Familiarity with at least one of the following: RAG systems, prompt engineering, vector databases, embeddings, or basic agent patterns.
- Solid foundation in software engineering basics including Git, REST APIs, JSON, SQL, and at least one cloud environment.
- Strong written and verbal communication skills with a willingness to ask questions and engage in technical discussion.
- Bachelor's degree in Computer Science, Data Science, Machine Learning, Engineering, or a related field, or equivalent demonstrable experience.
Preferred qualifications
- Experience with at least one AI framework such as LangChain, LlamaIndex, Hugging Face Transformers, or DSPy.
- Exposure to vector databases (Pinecone, Weaviate, pgvector, Vertex AI Vector Search) and embedding models.
- Familiarity with one major cloud platform (GCP, Azure, or AWS), particularly the managed AI services.
- Comfort with Docker, basic CI/CD workflows, and modern engineering practices.
- A portfolio of personal projects, open source contributions, hackathon work, or coursework that demonstrates curiosity and initiative in AI.
- Experience with web frameworks (FastAPI, Flask) or frontend basics (React, TypeScript) is a plus but not required.
- Coursework or self-directed learning in machine learning, deep learning, NLP, or information retrieval.
- Core Programming: Python (required), familiarity with JavaScript or TypeScript helpful.
- AI Tooling: Comfort calling LLM APIs, basic prompt engineering, familiarity with at least one framework (LangChain, LlamaIndex, Hugging Face).
- Data and Storage: SQL, JSON, basic familiarity with vector databases and traditional databases.
- Cloud and Engineering: Git, REST APIs, basic Docker, at least one cloud environment (GCP, Azure, or AWS).
- Bonus: Notebook environments (Jupyter, Colab), evaluation tools (RAGAS, LangSmith, Weights and Biases), basic frontend development.
- Curiosity: Genuine interest in how AI works and a habit of digging into details rather than treating models as black boxes.
- Ownership: Willingness to see problems through, even when the path is unclear.
- Communication: Comfort asking questions, sharing progress, and explaining your thinking.
- Quality Mindset: Pride in writing code that is clear, tested, and easy for others to work with.
- Learning Velocity: A track record of picking up new tools, languages, and concepts quickly.
- Collaboration: Generosity with teammates, openness to feedback, and willingness to help others succeed.
Tags & focus areas
Used for matching and alerts on DevFound Fulltime Ai Ai Engineer