Original job description
About RaiseDesk
RaiseDesk is building a fundraising operating platform for companies actively raising capital. We’re a small, fully remote team looking for senior people who work independently and ship.
About the role
We are looking for an Applied AI Engineer to own the intelligence layer of the platform: the data that powers investor discovery, the models and retrieval that rank and explain it, and the evaluation that proves any of it is working. This is an applied engineering role, not a research role. The work is building systems that run in production.
What you’ll do
Design and build the data pipelines that ingest, clean and organize third-party records at scale
Build entity resolution and record matching across inconsistent sources, preserving provenance
Design and ship LLM-powered features including retrieval, ranking, drafting, classification and summarization
Own evaluation end to end: build the harness, define the metrics, and measure whether changes actually improve quality
Manage latency, cost, guardrails and fallback behaviour for model-backed features
Use AI coding agents as a primary development tool, and take responsibility for reviewing and verifying what they produce
## Requirements
5+ years of professional software or data engineering experience, including recent hands-on applied AI work
Strong Python, and strong SQL against relational databases such as PostgreSQL or MySQL
Production experience with LLM applications: retrieval and RAG, system-level prompt design, tool calling, and structured output
Demonstrated ownership of evaluation frameworks. You can explain how you knew a change made the system better
Experience building data pipelines from scratch, including schema, ingestion, storage and monitoring
Daily working experience with AI coding agents, and a clear approach to reviewing and verifying their output
Demonstrated ability to work independently and manage your own scope and priorities
## Preferred
Experience with entity resolution or record linkage on messy real-world data
Background in search, ranking or recommendation systems
Experience with vector databases or embedding-based retrieval in production
Experience with workflow orchestration tools such as Airflow or Dagster
Familiarity with Go
## How we work
The team is small and entirely senior. There is no project manager and no formal onboarding program. You will be given context and outcomes, and you will decide how to get there. We expect engineers to scope their own work, manage their own time, and raise problems early. This suits people who learn quickly and independently, and who are comfortable making decisions without close direction.
Location
This role is fully remote. You must be legally entitled to work in Canada or the United States and perform the work from within one of these countries.
Apply
Apply through our careers page: https://raisedesk.io/careers/senior-applied-ai-engineer/apply
More about this job
Responsibilities
Build and operate data pipelines and entity-resolution systems that organize third-party records, and develop production LLM features for retrieval, ranking, drafting, classification, and summarization. Own end-to-end evaluation and production quality, including latency, cost, guardrails, and fallback behavior.
Requirements
Requires at least five years of software or data engineering experience, recent hands-on applied AI work, strong Python and SQL, and production experience with LLM applications and data pipelines. Candidates must have owned evaluation frameworks, use AI coding agents regularly and verify their output, and be able to work independently.
Skills
- Python
- SQL
- Applied AI
- Large Language Models
- Retrieval-Augmented Generation
- Prompt Design
- Tool Calling
- Structured Output
- Evaluation Frameworks
- Data Pipelines
- Entity Resolution
- Record Linkage
- Search and Ranking
- Vector Databases
- AI Coding Agents
- PostgreSQL
Remote locations (standardized)
Visa sponsorship
Not detected in the job text
Categories
- Technology
- Software
- Data & Analytics
- Engineering
Keywords
- Python
- SQL
- PostgreSQL
- MySQL
- Large Language Models
- Retrieval-Augmented Generation
- Retrieval
- Ranking
- Drafting
- Classification
- Summarization
- Prompt Design
- Tool Calling
- Structured Output
- Evaluation Frameworks
- Data Pipelines
- Entity Resolution
- Record Matching
- Data Provenance
- Vector Databases
- Embedding-Based Retrieval
- Search Systems
- Recommendation Systems
- Airflow
- Dagster
- Go
- AI Coding Agents
- Latency Management
- Guardrails
- Fundraising
