AI Expert for ATS CV Screening Module

AI/LLM Engineer – Production AI Screening Engine for ATS About Us We are building a next-generation Applicant Tracking System (ATS) that leverages AI to automate candidate screening, evaluation, and ranking. Our ATS platform, backend, frontend, and databases are already production-ready. We are not looking for someone to build an ATS. We are looking for an experienced AI/LLM Engineer to build a production-grade AI Screening Engine that integrates with our existing ATS through REST APIs. The goal is to achieve high-quality candidate evaluations while keeping AI costs as low as possible. Project Overview The selected engineer will design and implement an AI service responsible for evaluating candidates using one or multiple Large Language Models. The service will receive candidate information from our ATS, intelligently process it, and return structured evaluation results in JSON format. We expect the solution to be reliable, scalable, deterministic, and optimized for production use. Existing System We already have: Production ATS Backend APIs Candidate workflows CV parsing Candidate database Existing AI prompts Production infrastructure The AI service will simply consume our data and return structured evaluation results. Responsibilities Design and improve production-grade prompts. Redesign our current prompt architecture if necessary. Build a Python REST API (FastAPI preferred). Integrate multiple LLM providers (OpenAI, Gemini, DeepSeek, etc.). Recommend the most appropriate model for each task. Optimize prompts for: accuracy consistency reasoning quality structured outputs low hallucination rate predictable behavior Design robust JSON output schemas. Implement validation and error handling. Compare different AI models based on: quality latency token usage operational cost Recommend a multi-model strategy that balances quality and cost. Help integrate the AI service into our ATS. Technical Requirements Required: Strong Prompt Engineering experience Production experience with LLMs Python FastAPI REST APIs Pydantic JSON Schema OpenAI API Google Gemini API DeepSeek API Git We are not looking for simple prompt writing. We are looking for someone who understands how to build reliable AI systems, including: Prompt architecture Model orchestration Structured outputs Validation Cost optimization Model benchmarking Reliability Production deployment Deliverables The selected engineer will deliver: Production-ready prompt templates AI evaluation workflow Python REST API Structured JSON schemas Multi-model integration Model selection strategy Prompt versioning Validation pipeline Documentation Integration guide Cost Optimization A major objective of this project is to achieve the best possible evaluation quality while minimizing AI costs. The proposed solution should intelligently balance: evaluation accuracy reasoning capability token consumption API cost latency scalability We are open to hybrid approaches that use different models for different stages of the evaluation pipeline.

Project ID: 40547286