About the role:
You will build the backend AI + data infrastructure that powers our next-generation talent and evaluation platform. This includes work across data modeling, embeddings, feature extraction, pipeline creation, and serving AI-powered insights at scale.

You’ll partner closely with engineering and product to design systems that process structured and unstructured data, generate semantic representations, and expose intelligence through reliable APIs.

This is a hands-on engineering role focused on scalable AI infrastructure — not research or prototyping.

Responsibilities

AI + Data Infrastructure

• Design and maintain data models for structured + unstructured data

• Build data ingestion and processing pipelines

• Implement embedding pipelines and vector search systems

• Develop feature extraction logic using LLMs and smaller models

• Optimize for accuracy, latency, and cost
  
  

Platform Engineering

• Design and implement backend services that expose AI-driven insights

• Maintain versioning, evaluation checks, and health monitoring for AI outputs

• Optimize performance, caching, and retrieval logic for semantic data

• Collaborate with engineers to ensure data consistency and reliability
  
  

Model Integration

• Integrate LLM APIs, embedding models, and small LMs

• Build evaluation harnesses to validate extraction quality

• Monitor drift, degradation, or inconsistencies over time
  
  

General

• Write maintainable backend code (Node/Python/Go)

• Work cross-functionally with product and engineering

• Ensure AI systems are robust, observable, and production-ready
  
  

Required Skills

• Strong backend engineering experience

• Experience designing or maintaining data pipelines

• Practical experience with embeddings and vector databases

• Familiarity with LLM application patterns (prompting, extraction, RAG)

• Strong SQL + NoSQL fundamentals

• Ability to design scalable APIs

• Understanding of evaluation metrics for AI systems

Nice to Have

• Experience with LangChain or LlamaIndex

• Experience with smaller open-source models (Llama/Qwen/Mistral/etc.)

• Experience with ETL frameworks (Airflow, Dagster, Temporal)

• Interest in skills-based matching / intelligence systems (not required)

Success Indicators

• AI pipelines are stable, observable, and scalable

• Data processing is efficient and correct

• API services return reliable and consistent insights

• Embedding systems improve retrieval quality over time