Most recruiting tools were built for humans clicking buttons. You log in, post a job, wait for applications, open a spreadsheet, and start reading resumes one by one. It works, technically. But if you're building AI-native products or running a fast-moving startup, that workflow is a bottleneck.

The Noxx Agentic API is built for a different model. It lets AI agents, internal tools, and automated systems handle job posting, candidate screening, and interview scheduling programmatically. No dashboard required. No manual steps in the middle.

This guide walks you through how to build that workflow — from authentication to a fully automated hiring loop.

Why Recruiting Needs a Programmatic Layer

If you're a technical founder or ops lead, you've probably already automated large parts of your business. Your CRM updates automatically. Your onboarding flows trigger on conditions. Your infrastructure scales without someone clicking a button.

Recruiting is usually the exception. It still runs on manual reviews, email chains, and agency relationships that charge 20–30% of a candidate's annual salary whether the hire works out or not.

A recruiting API integration changes that. Instead of treating hiring as a one-off manual process, you can wire it into your existing systems. When headcount is approved, a job posts automatically. When candidates are screened, results flow into your internal tools. When you're ready to interview, scheduling happens without back-and-forth email.

That's the model the Noxx Agentic API is designed for.

What the Noxx Agentic API Actually Does

Noxx automates end-to-end recruiting: you provide a job, the AI screens 1,000+ candidates using 40+ evaluation signals, and surfaces the top 10 matches within 7 days. The Agentic API exposes that entire process programmatically.

Concretely, the API supports:

• Job posting — submit a job description, role requirements, location preferences, salary range, and time zone filters via a structured request

• Candidate screening — Noxx's AI runs the 40+ signal evaluation against the candidate pool and returns ranked results

• Candidate retrieval — pull the top matches, including AI-generated match scores, skill assessments, and salary expectations

• Interview scheduling — trigger interview scheduling for selected candidates without leaving your system

This makes Noxx useful not just as a standalone product, but as a recruiting layer inside a larger AI agent stack. If you're building an internal hiring agent, a workforce planning tool, or an automated ops system, you can connect it directly to Noxx's infrastructure.

Core Workflow: Four Stages of Programmatic Hiring

Here's how a full recruiting workflow looks when built on the Noxx Agentic API.

Stage 1: Authentication and Setup

Before any API calls, you'll authenticate with your Noxx credentials to get an access token. Standard bearer token authentication — nothing unusual here.

The response returns a token you'll include in the Authorization header for all subsequent requests. Tokens are scoped to your account and the roles you've configured.

Keep your credentials in environment variables, not hardcoded in your codebase. If you're building an agent that runs this workflow autonomously, store them in your secrets manager and rotate on a schedule.

Stage 2: Posting a Job Programmatically

Once authenticated, posting a job is a single API call. You send a structured payload with the role details, and Noxx begins the screening process immediately.

The regions field maps to Noxx's global talent network. You can target LATAM, Southeast Asia, Eastern Europe, or leave it open for global sourcing. The timezone_overlap parameter filters candidates who can work within your team's hours — a small detail that saves a lot of pain later.

The response includes a job_id you'll use to poll for results.

Stage 3: Retrieving Screened Candidates

Noxx runs the AI screening in the background. Within 7 days (often faster), the top 10 candidates are ready. You retrieve them with a GET request against your job ID.

The response returns a ranked list of candidates. Each candidate object includes:

• Match score — a composite score based on the 40+ AI evaluation signals

• Skills assessment — how the candidate maps to your stated requirements

• Salary expectations — what the candidate is looking for, so there are no surprises

• Availability — when they can start

• AI-generated interview questions — role-specific questions Noxx generates based on the candidate's profile and your job requirements

That last point matters. You're not just getting a list of names. You're getting a ready-to-use interview kit for each candidate, generated by the same AI that screened them.

A sample candidate object looks something like this:

You can pipe this directly into your internal tooling — a Notion database, a Slack message, a custom dashboard, or whatever your team uses to manage hiring.

Stage 4: Scheduling Interviews via the API

When you've reviewed the candidates and want to move forward, you trigger interview scheduling through the API. No email threads, no calendar back-and-forth.

Noxx handles the coordination and confirms a time that works for both sides. The candidate gets notified, and your system receives a confirmation event you can use to trigger downstream steps — calendar invites, prep materials, internal notifications.

Building an Agentic Hiring Loop

The four stages above work as individual API calls. But the real value comes when you chain them into an autonomous agent loop.

Here's a simple architecture that works well for seed-stage teams:

Trigger: Headcount approval in your internal system (a Notion status change, a Jira ticket, a webhook from your HR tool)

Step 1: Agent reads the approved role details and calls POST /jobs to create the job in Noxx

Step 2: Agent polls GET /jobs/{job_id}/candidates on a schedule (or listens for a webhook) until results are ready

Step 3: Agent formats the top candidates and posts a summary to a Slack channel or internal dashboard for the hiring manager to review

Step 4: Hiring manager marks candidates they want to interview (a simple button click or status update in your tool)

Step 5: Agent calls POST /schedule for each selected candidate and adds confirmed interviews to the team calendar

Step 6: If a hire is made, the 3% fee is triggered. If no hire, no cost.

This loop removes every manual step except the actual hiring decision. The hiring manager still chooses who to interview and who to hire. The agent handles everything else.

For teams building more sophisticated systems, you can extend this further. Connect Noxx's candidate data to your internal scoring models. Build feedback loops that improve your job descriptions over time. Trigger re-posts automatically if a role doesn't fill within a defined window.

The API is designed to fit inside larger agent architectures, not just replace a recruiter dashboard.

When to Use the API vs. the Standard Interface

Not every team needs the API. Here's a quick way to think about it:

The standard interface is fast and requires no setup. The API is for teams that want recruiting wired into their existing systems or agent workflows.

Both use the same underlying AI screening engine. Both deliver the same top 10 candidates in 7 days. Both charge the same 3% success fee — only if you hire.

Pricing Still Works the Same Way

This is worth stating clearly because it's different from how most APIs work.

You don't pay per API call. You don't pay a monthly subscription for API access. The pricing model is identical to the standard product: 3% of the hired candidate's annual salary, paid only if you make a hire.

That means you can build an entire automated recruiting workflow, run it across multiple open roles, and pay nothing until someone gets hired. No upfront cost, no credit card required to get started.

Compare that to traditional recruiter fees of 20–30% of annual salary, paid regardless of outcome, and the math is straightforward. A $100,000 hire costs $3,000 through Noxx. The same hire through a traditional agency could cost $20,000–$30,000.

The API doesn't change that equation. It just makes the process faster and more automated.

FAQs

What programming languages work with the Noxx Agentic API?
The API is REST-based and returns JSON, so it works with any language that can make HTTP requests — Python, JavaScript/Node, Go, Ruby, and others. There's no proprietary SDK required, though you can build one around the endpoints if you want to standardize how your internal tools interact with it.

How long does it take to get candidates back after posting a job via the API?
The same timeline as the standard product: top 10 candidates within 7 days. In practice, results often come back faster depending on the role and region you're targeting.

Can I filter candidates by region and time zone through the API?
Yes. The job posting endpoint accepts regions (LATAM, Southeast Asia, Eastern Europe, or global) and timezone_overlap parameters. You can also specify salary range to filter candidates whose expectations match your budget.

Does the API support webhook notifications, or do I need to poll for results?
The API supports polling via the candidates endpoint. For production agentic workflows, check the Noxx documentation at noxx.ai for the latest on event-based notifications, as webhook support may be available depending on your account configuration.

What happens if none of the top 10 candidates are a fit?
You pay nothing. The 3% fee only applies when you make a hire. If the first batch doesn't work, you can repost with adjusted requirements or filters and Noxx runs the screening again.

Can multiple AI agents use the same Noxx API credentials?
Credentials are scoped to your account. If you're running multiple agents or services that need to post jobs and retrieve candidates, you can use the same credentials across them. For production systems with multiple services, it's good practice to use separate credentials per service where possible — check the Noxx documentation for account-level configuration options.

Is the Noxx Agentic API suitable for high-volume hiring, or is it designed for startup-scale use?
It's built for startup-scale hiring — typically the first 5–20 engineers or operators at seed-to-Series A companies. If you're hiring across multiple roles simultaneously, the API handles that well. For enterprise-scale volume hiring (hundreds of roles), the product is optimized differently than what you'd find in enterprise platforms like HireVue or Kula.

Start Building

The Noxx Agentic API gives you a recruiting layer you can wire into any system. Post jobs programmatically, retrieve AI-screened candidates, schedule interviews automatically, and pay 3% only when you hire.

If you're building an internal hiring agent or want to connect recruiting to your existing ops stack, this is the most direct path to doing it without paying traditional recruiter fees or buying a subscription you may not fully use.

Learn more and get started at noxx.ai.