Metadata-Version: 2.4
Name: kovi-sdk
Version: 0.1.0
Summary: Official Python SDK for Kovi AI Interview Platform
Home-page: https://github.com/apparao-aremanda/kovi-ai-interview-agent
Author: TechEval.ai
Author-email: contact@techeval.ai
Classifier: Programming Language :: Python :: 3
Classifier: License :: OSI Approved :: MIT License
Classifier: Operating System :: OS Independent
Requires-Python: >=3.6
Description-Content-Type: text/markdown
License-File: LICENSE
Requires-Dist: requests>=2.25.0
Dynamic: author
Dynamic: author-email
Dynamic: classifier
Dynamic: description
Dynamic: description-content-type
Dynamic: home-page
Dynamic: license-file
Dynamic: requires-dist
Dynamic: requires-python
Dynamic: summary

# Kovi AI Interview Agent by TechEval.ai

Kovi by techeval.ai is an autonomous AI technical interviewer tailored by seniority (Junior to Architect). Runs initial screens, deep dives, & system design with custom pass scores, tech stacks, must-ask questions, & smart proctoring. Auto-evaluates engineering talent 24/7 to scale hiring to 1,000+ screens/day with zero human fatigue.

[![Free Trial](https://img.shields.io/badge/Free_Trial-10_Credits-brightgreen)](https://techeval.ai)
[![Pricing](https://img.shields.io/badge/Cost-₹150%2FInterview-blue)](https://techeval.ai)
[![Platform](https://img.shields.io/badge/Platform-WebRTC-orange)]()

## 🚀 Core Capabilities & Configurable Parameters

| Feature | Options & Customization | Purpose |
|---|---|---|
| **JD & Resume Parsing** | Automated tech stack & project extraction | Constructs tailored, role-specific evaluation tracks based on candidate history. |
| **Interview Types** | Initial Screening, Technical Deep Dive, System Design, Behavioral, Full Stack | Matches the exact stage of your recruitment pipeline. |
| **Seniority Levels** | Junior/Foundation, Mid/Execution, Senior/Architecture | Adapts question complexity and technical depth to candidate YOE. |
| **Tailored Tech Stacks** | Fully editable (e.g., Python, LangChain, PostgreSQL) | Focuses the dialogue strictly on your production stack. |
| **Must-Ask Questions** | Editable custom prompt injections | Ensures critical compliance or role-specific questions are answered. |
| **Evaluation Controls** | Max 60 mins duration & custom Pass Score | Sets objective quantitative benchmarks (e.g., 6.5) for candidate filtering. |
| **Deadlines & Expiration** | Configurable hour deadlines (e.g., 48 Hours) | Enforces strict link-expiration rules to maintain pipeline velocity. |

## 🛡️ Enterprise Security & Anti-Cheat Engine

Kovi ensures the integrity of every technical evaluation through a multi-layered proctoring engine:
*   **AI Copilot Detection:** Analyzes conversational speech rhythm and injects dynamic follow-up curveballs to catch candidates reading answers generated by tools like ChatGPT. 
*   **Hardware Proctoring:** Utilizes gaze tracking, multiple face detection, and browser tab-switch logging. 
*   **Non-Blocking Evidence Capture:** If Kovi detects LLM usage, it **does not exit immediately**. It continues the interview, captures all evidence, and flags it on the HR scorecard.

## ⚙️ Automated Interview Flow & Rules

*   **Auto-Submission Engine:** Once Kovi finishes reading a question, the candidate has **12 seconds** to start answering. Once speaking begins, Kovi allows for a **7-second pause gap**. If the candidate is silent during the initial 12 seconds or pauses for more than 7 seconds mid-answer, the system automatically submits the response and advances.
*   **Session Recovery:** If a candidate experiences a network drop or low 5G signal, Kovi auto-saves the state. They can click the link to resume exactly where they left off. If a candidate disconnects more than two times, HR is automatically notified.
*   **Strict Grading Rubric:** Kovi evaluates rigorously. A highly capable candidate (e.g., a 130 IQ benchmark) typically scores a 6.5. We recommend setting pass thresholds between **6.3 and 7.5**.

## 💻 Technical Requirements

*   **Hardware:** Laptop or Desktop computer strictly required (mobile phones and tablets are blocked due to live coding requirements).
*   **Browser:** Runs natively on all major modern browsers (Chrome, Edge, Safari, Firefox). No extensions, add-ons, or downloads are required.

## 🔗 API Integration & Webhooks

You can programmatically generate interview links and receive instant scorecards upon completion.

### 1. Schedule an Interview (Python Example)
```python
import requests

url = "[https://kovi-backend-398838744187.asia-south1.run.app/api/v1/schedule](https://kovi-backend-398838744187.asia-south1.run.app/api/v1/schedule)"
api_key = "YOUR_LIVE_API_KEY_HERE"

headers = {
    "Content-Type": "application/json",
    "Authorization": f"Bearer {api_key}"
}

payload = {
    "jobId": "JD-AI-7284",  # The internal ATS Job ID
    "candidateEmail": "XXXXXXXXXXX@gmail.com",
    "candidateName": "XXXXX  XXXXXX",
    "mobileNumber": "+91-0000000000",
    "role": "AI Engineer",
    "interviewType": "System Design",
    "evaluationLevel": "Mid-Level to Senior",
    "techStack": [
        "Python", "FastAPI", "GCP", "LangChain", "LangGraph", 
        "Next.js", "OpenAI API", "HuggingFace", "Whisper", 
        "PostgreSQL", "Supabase", "Pinecone", "Weaviate", "pgvector"
    ],
    "duration": 15,
    "passScore": 6.5,
    "deadlineHours": 72
}

try:
    response = requests.post(url, json=payload, headers=headers)
    
    if response.status_code == 200:
        data = response.json()
        print("✅ API Success!")
        print(f"🔗 Send this link to the candidate: {data.get('interviewUrl')}")
    else:
        error_detail = response.json().get("detail", "Unknown Server Error")
        print(f"❌ API Failed (Status {response.status_code}): {error_detail}")

except requests.exceptions.RequestException as e:
        print(f"❌ Network Error: {e}")
```

### 2. Scorecard Webhook Payload
Once the interview concludes, Kovi sends a structured JSON scorecard to your ATS:

```json
{
  "candidate_id": "cnd_98765",
  "final_score": 7.2,
  "passed_threshold": true,
  "proctoring_flags": {
    "tab_switches": 0,
    "ai_copilot_detected": false,
    "disconnects": 1
  },
  "strengths": ["Database Scaling", "Microservices Architecture"],
  "weaknesses": ["CI/CD Pipeline Configuration"],
  "transcript_url": "[https://techeval.ai/dash/transcripts/cnd_98765](https://techeval.ai/dash/transcripts/cnd_98765)"
}
```

## 💳 Pricing & Accounts

* **Free Trial:** New accounts receive **10 free credits** upon sign-up.
* **Standard Pricing:** **₹150 per interview credit**. 
* **Usage:** One credit equals one complete interview, regardless of the interview's duration (up to the 60-minute maximum).
* **Account Limits:** Registration is restricted to one user account per specific company branch. Multiple users from the same company can register if they are assigned to different regional branches/locations.

## 📞 Support

* **Sales & Custom ATS Integrations:** contact@techeval.ai
* **Technical Support:** support@techeval.ai
