Applied AI Engineer

Salary: $200,000 - 246,000 per year Requirements:

• 5+ years of professional software engineering experience, including ownership of at least one production system from design through operation.
• At least one year responsible for a production LLM-based system with real users, including post-launch feedback and iteration.
• Production experience with retrieval-augmented generation and measurable improvements to retrieval quality.
• Experience with structured outputs, prompt design, and evaluation of generated text.
• Strong Python skills and experience with PostgreSQL.
• Experience measuring the impact of model and prompt changes before deployment.
• Strong documentation and communication skills.
• Preferred: Experience with pgvector, fine-tuning open-weight models using LoRA or similar methods, serving models with vLLM on NVIDIA hardware, classification or entity-linking systems, or public records, regulatory filings, or civic data.

• Build and maintain Arthurs retrieval layer, including chunking, embeddings, PostgreSQL indexing with pgvector, reranking, and citations for generated text.
• Develop the generation pipeline for research analyses, including prompts, structured output schemas, recommended next steps, and fixed sets of client questions.
• Create classifiers for signal type, impact, and stakeholder stance, starting with prompting and using fine-tuned models when volume warrants it.
• Develop entity extraction and linking to connect names in incoming documents with existing stakeholder records.
• Maintain an evaluation set based on past projects and run it against every prompt or model change.
• Prepare analyst-approved briefs as training data and run fine-tuning jobs on our hardware.
• Work with analysts who review briefs to identify recurring corrections and translate them into system improvements.
• Document prompts, schemas, evaluation results, and operating procedures.

• AI
• CTO
• Fine-tuning
• Hardware
• LLM
• PostgreSQL
• Python
• vLLM

More:

At Eve & Co, we have spent more than a decade helping developers of large infrastructure projects understand the communities, agencies, and politics involved. Arthur is our software product, built on that experience: it gathers information such as public meeting agendas, regulatory filings, local news, community petitions, and field notes, connects those signals to communities, and provides clients with analytics and insights. Our small product team works within an established firm, bringing analysts and engineers together. We are hiring an Applied AI Engineer to lead the retrieval, generation, and evaluation systems behind Arthurs briefs; our models run on hardware we control, and every generated sentence must be traceable to a source record. The role reports to Arthurs CEO or CTO and works with engineers, a designer, and analysts who prepare and review data. This is a full-time, onsite position in New York City, with a base salary of $200,000–$246,000 plus equity, depending on experience. Benefits include a 401(k) and matching, dental, health, life, and vision insurance, paid time off, parental leave, professional development assistance, and a retirement plan. This is a direct-hire position; agencies, third-party recruiters, contract-to-contract, and contract-to-hire arrangements are not accepted.

last updated 39 week of 2026