An invisible AI overlay that lives on your screen — answers questions, solves problems, analyzes screenshots, and feeds you real-time intelligence. Works during interviews, exams, meetings, or anything on your screen. Invisible to screen recordings. Undetectable by proctoring software. You see everything. They see nothing.

👻 Invisible by Design — Screen capture protected. Hidden from taskbar. Undetectable by Zoom, Teams, Google Meet, and every proctoring tool on the market.

💸 $0. Forever. — No subscriptions. No credits. No paywalls. Bring your own free API keys and own it.

⚡ Stupidly Fast — Powered by Cerebras (~2,000-3,000 tokens/sec) and Groq (~400-700 tokens/sec) — the two fastest AI inference engines on Earth. Answers hit your screen before the interviewer finishes talking. Nothing else comes close.

Aura isn't just another interview prep tool. It's your secret weapon for any challenge—from aptitude tests to quantitative brain-twisters, behavioral showdowns to certification exams. Aura is a revolutionary AI assistant that operates in real-time, providing candidates with the critical insights they need to excel in high-stakes situations—all without ever tripping tab-switching warnings, thanks to its stealthy, seamless design.

Imagine having a world-class expert whispering in your ear, helping you deconstruct complex problems, articulate your thoughts, and navigate the toughest questions with confidence. That's Aura.

Aura sits on top of your screen as an invisible overlay during live interviews and exams. It listens to conversations in real-time via Deepgram speech-to-text, generates intelligent coaching responses using blazing-fast LLMs (Groq, Cerebras, Gemini), and lets you screenshot any problem for instant Vision AI solutions—all while staying completely hidden from screen recordings, proctoring software, and human observers.

Real screenshots and recordings from a live Aura session — no mockups, no staging.

Step 1 · Your Profile — Enter your details and pick an interview focus

Fill in your name, target company, and role, then select your interview focus (Behavioral, Coding, System Design, Data Structures, Cultural Fit, or HR/Screening). Paste your full resume and job description for hyper-personalized coaching. Hit the ⚡ Demo button to auto-fill with sample data for quick testing.

Step 2 · AI Setup — Configure primary, secondary, and vision models

Choose your Primary AI Model (default for all responses — Cerebras shown here for maximum speed) and an optional Secondary Model (switch anytime with Alt+W). Below, configure your Vision AI Model for screenshot analysis — Gemini delivers the best accuracy, Groq Llama 4 is the fastest. The Quick Switch Hotkeys panel shows Alt+Q, Alt+W, Alt+E for instant model switching mid-interview.

Step 3 · Advanced Config — Manage API keys and provider connections

The Advanced Config tab is the raw control panel. Enter your Deepgram API key (speech-to-text), then manage multiple API keys per provider — notice Groq showing 5 keys for maximum rate-limit resilience. Each provider shows its base URL, keys, and a Test Connection button to verify everything works before going live. Changes are saved directly to ai_providers.json and .env.

Step 4 · Commands & Tips — Complete hotkey reference with pro tips

The final onboarding tab is your cheat sheet. Three highlighted pro-tip cards cover: 🔒 Stealth Professional Tip (always use global hotkeys — never click Aura during proctored exams), 📸 Multi-Screenshot Analysis (queue up to 4 screenshots with Alt+S, process all at once with Alt+P), and ⚡ Speed Strategy (use Alt+E to auto-select the fastest healthy provider). Below, every hotkey is organized by category with color-coded badges.

Microphone Permission Request — Aura's Pre-Flight Check runs before every session. It validates your microphone access, backend connection, Deepgram API, and all configured AI providers. Grant the mic permission prompt to proceed.

All Systems Go ✅ — Every check is green: Microphone Permission OK, Microphone Selected (NVIDIA Broadcast detected), Backend Connected, Deepgram API OK, Primary Cerebras OK, Secondary Groq OK, Primary Vision Gemini OK, Secondary Vision Groq OK. Hit Start Interview and you're live.

See how Aura overlays directly on top of LeetCode during a live coding session — invisible to screen recordings, fully transparent, and controlled entirely via hotkeys.

Screen Share Protection — Aura runs during a live LeetCode session with "Share Entire Screen" active. The Aura window overlays the coding environment but is excluded from the screen capture via WDA_EXCLUDEFROMCAPTURE. Notice the "Microphone Muted" status bar and model switcher visible only to the user.

Maximum Transparency Mode — With Alt+1 (40% opacity), Aura becomes a near-invisible ghost overlay. The Live Interview Session indicator, MUTED/RESET controls, and AI model badges (Primary: Cerebras) are barely visible — just enough for you, completely invisible to anyone observing your screen. The LeetCode problem beneath is fully readable.

Vision AI + Screen Share — When using Vision AI during a proctored session, Aura shares your entire screen for screenshot capture. The "Also share system audio" toggle (circled) enables audio capture alongside visual analysis. The Aura window itself remains hidden from the capture feed — only you see it.

Note: Click any link below to watch the .mp4 demo directly. These recordings show actual Aura features running in real-time.

Clock's ticking. Stakes are high. Aura's got you covered—whether it's an interview or an exam. Its Stealth Mode ensures you can access its powerful features without raising any red flags, keeping your focus entirely on the task at hand.

The Curveball: "Tell me about a time you turned failure into success."

• Without Aura: Stammering, scrambling, stuck.
• With Aura (Stealth Mode active):
  • Real-time transcription catches every word, discreetly.
  • STAR-method gold pops up on your screen, invisible to others—customized to your resume and role.
  • You drop a story so smooth, they're taking notes.

Boom: You're unforgettable.

The Brainteaser: "Calculate the probability of 7 consecutive heads in all three coins tossed at same time in 100 coin tosses."

• With Aura (Stealth Mode is your superpower here):
  1. Alt+Shift+S → Proctoring Stealth Mode activates. Aura becomes a transparent, click-through overlay, invisible to screen recording and proctoring software.
  2. Alt+V → Vision Mode activates, ready to analyze.
  3. Alt+S → Silently screenshots the problem from the exam platform.
  4. Alt+P → Quantitative AI analyzes the problem in the background.
  5. Solutions appear ghosted on your screen—only you can see them.
  6. Submit answers confidently without ever switching tabs or alerting proctoring systems.
  7. Control Aura's window movements and scroll through suggestions using global hotkeys, all while the exam window remains active and focused.

You solve complex problems with ease, leaving no trace.

The Puzzle: Reverse-engineer a slow API call under pressure during a live coding session.

• Without Aura: Sweat-soaked guesswork.
• With Aura (Stealth Mode ensuring seamless assistance):
  1. Alt+V → Vision Mode ignites, ready for the code.
  2. Alt+S → Screenshot the problematic code or error message, discreetly.
  3. Alt+P → Vision AI unleashes its power, overlaying suggestions:
     • Killer solutions in JS, Python, Java, SQL, and more.
     • Time complexity breakdowns for optimal answers.
     • Pro-level explanations to articulate your thought process.
     • All assistance is delivered via Aura's overlay, invisible to screen sharing.

You code like a rockstar, navigating complex challenges with AI-powered insights, leaving them stunned.

The Beast: "Design a global payment system for millions."

• With Aura (Stealth Mode providing discreet guidance):
  • Real-time transcription catches the requirements as the interviewer speaks.
  • Aura provides quick analysis of constraints, goals, and trade-offs—visible only to you.
  • API Design: RESTful/GraphQL endpoint suggestions with detailed descriptions.
  • Scalable architectures—sharding, replication, microservices, oh my!
  • Caching tricks and load balancing hacks.
  • Buzzwords like Kafka, Redis, Kubernetes—served hot and contextually.

You sketch a masterpiece, guided by expert insights, and they're hiring you yesterday.

Aura doesn't just compete—it crushes subscription traps like Interview Coder, Parakeet AI, LockedIn AI and similars. Check the carnage:

Aura's Knockout Punch: Fastest inference in the game (Cerebras + Groq), totally invisible (screen-capture-proof stealth), and $0 forever. Competitors charge $25/month for slower, weaker, detectable tools. Aura buries them all.

Aura's Stealth Mode is engineered to provide powerful AI assistance while remaining completely invisible and undetectable to proctoring software, screen recording tools, and human observers. This is crucial for high-stakes online exams, remote interviews, and any situation requiring discreet support.

What it is: Stealth Mode transforms Aura into a transparent, click-through overlay that seamlessly integrates with your workflow. It operates without creating new windows that could trigger alerts, and all interactions are managed via global hotkeys—meaning you never need to switch focus from your primary application.

Core Components:

📖 Full stealth guide: PROCTORING_STEALTH_GUIDE.md

When combined with Stealth Mode, Vision AI becomes even more potent:

• Multi-Screenshot Queue — Capture up to 4 screenshots with Alt+S, building a queue of problems. Process the entire batch at once with Alt+P — perfect for multi-part exam questions or complex code spanning multiple files
• Auto Content-Type Detection — Aura automatically identifies whether a screenshot contains code, a diagram, a math problem, a database schema, or plain text — and adapts its analysis strategy accordingly
• Instant Problem Analysis — Screenshot questions from any platform (LeetCode, HackerRank, CodeSignal, Google Docs, PDF exams), process them, get solutions overlaid invisibly
• Multi-Language Code Solutions — Python, Java, C++, JavaScript, TypeScript, SQL, Go, Rust, C#, and more — with syntax-highlighted answers
• System Architecture Understanding — Analyzes architecture diagrams, UML, flowcharts — suggests optimizations, identifies bottlenecks, recommends design patterns
• Database Schema Analysis — Understands ER diagrams, table relationships, generates optimal queries, suggests indexing strategies
• Quantitative Problem Solving — Breaks down complex math, probability, statistics, logic puzzles, and financial modeling problems step-by-step
• Multiple Solution Approaches — Every problem gets multiple approaches with time/space complexity analysis, trade-off comparisons, and interview presentation tips
• Vision Model Cycling — Switch between vision providers on the fly with Alt+T (e.g., Gemini for accuracy, Groq Llama 4 for speed)
• Queue Management — Clear your screenshot queue with Alt+R to start fresh between problems

Best vision setup: Use Gemini (free, excellent accuracy) as primary + Groq Llama 4 (fastest) as backup. Switch between them with Alt+T.

• Primary + Secondary Models — Run two different AIs simultaneously (e.g., Cerebras for speed + Groq for vision). All processing happens invisibly
• Instant Model Switching — Alt+Q (primary), Alt+W (secondary), Alt+E (auto-select fastest healthy provider) — swap AI strategies mid-interview without leaving your active window
• Auto-Failover System — If one provider hits rate limits, Aura auto-switches to the next key or provider. Zero downtime, zero manual intervention
• API Key Rotation — Add multiple free keys per provider via apiKeys array — Aura round-robins between them, multiplying your effective throughput
• Multi-Provider Support — Cerebras (fastest inference), Groq (fast + vision), Gemini (best vision accuracy), OpenRouter (gateway to 100+ models) — all free tiers
• Health Monitoring — Aura continuously monitors provider health and response times, auto-selecting the fastest available provider when you press Alt+E
• Configurable Per-Session — Choose different provider combinations for different interview types (e.g., Cerebras for behavioral, Gemini for system design with diagrams)

• Live Transcription — Deepgram-powered, high-accuracy speech-to-text with continuous streaming. Captures both interviewer questions and your responses in real-time
• Context-Aware Coaching — Aura knows your resume, job description, target role, and the full conversation history. Every response is hyper-personalized to you and the specific question being asked
• Conversation Memory — Remembers the entire interview (configurable up to 20 exchanges) for cross-referencing past questions, detecting follow-ups, and maintaining consistent narrative
• Candidate Response Tracking — Optionally tracks what you say so Aura can provide follow-up suggestions, catch mistakes, and help you build on previous answers
• Smart Audio Controls — Alt+M to toggle mic mute, Alt+U for universal pause/resume of all AI processing — all without touching the mouse
• Auto-Reconnect — If the Deepgram connection drops, Aura automatically reconnects and resumes transcription seamlessly
• Full or Brief Answers — Toggle between detailed, interview-ready responses and quick bullet-point hints via GENERATE_FULL_ANSWERS in .env

This diagram illustrates Aura's multi-layered architecture—designed for real-time performance, intelligence, and unparalleled discretion through its Stealth Mode capabilities.

The "Stealth & Privacy" layer in the OS Interaction component is key to Aura's undetectable nature.

Never leave the interview window. Control everything instantly with our ergonomic hotkey system, designed for maximum discretion and efficiency. All hotkeys function seamlessly while Aura is in Stealth Mode.

Complete control. Zero disruption. Maximum advantage. All under the cloak of Stealth Mode.

💡 Scroll Speed is configurable via SCROLL_SPEED_PX and SCROLL_INTERVAL_MS in your .env file. See Configuration for preset combos.

git clone https://github.com/Rkcr7/Aura-AI
cd Aura-AI

click run.bat to auto install depedencies
OR
# Create & activate virtual environment
python -m venv venv
venv\Scripts\activate

# Install all dependencies
pip install -r requirements.txt

You need two types of keys: one for speech-to-text (Deepgram) and one or more for AI models (Groq, Cerebras, Gemini). All providers offer generous free tiers — no credit card required.

📖 Full guide: Obtaining API Keys

The app auto-creates .env from .env.example on first launch. Just add your Deepgram key:

cp .env.example .env

DEEPGRAM_API_KEY="your_deepgram_api_key_here"

cp ai_providers.example.json ai_providers.json

Open ai_providers.json and replace placeholder keys with your actual keys. You don't need all 4 providers — even a single provider with one key works fine.

Option A — Direct launch:

python main.py

Option B — One-click automated launcher (recommended):

run.bat

run.bat automatically checks Python is installed, creates the virtual environment, installs/updates all dependencies, and launches the app — all in one double-click. No terminal commands needed.

Option C — Silent/invisible launch (maximum stealth):

silent_run.vbs

Double-click silent_run.vbs to launch Aura with zero visible terminal windows. The entire startup process runs completely hidden — perfect for stealth situations.

1. Profile — Enter your name, target company, role
2. Resume — Paste your full resume (unlimited length, no truncation)
3. Job Description — Paste the JD for tailored coaching
4. AI Models — Select primary and secondary LLM providers
5. Vision (optional) — Choose a vision model for screenshot analysis
6. Start — Hit "Start Interview" and you're live!

💡 Tip: Click the ⚡ Demo button on the profile tab to instantly fill the form with sample data for quick testing.

Every provider Aura supports offers free tiers that are more than enough for interview and exam use. No credit card required for any of them.

Deepgram is the core engine that powers Aura's real-time transcription. It listens to your microphone and the interviewer's audio, converting speech to text that gets fed to the AI for instant coaching responses. Without Deepgram, Aura can't hear the interview.

Cerebras is the fastest inference provider on the planet. Responses come back almost instantly. Ideal as your primary model for text-based coaching.

Groq provides extremely fast inference with both text and vision model support. Great as a secondary model or for Vision AI.

Google's Gemini models deliver excellent Vision AI accuracy—best for analyzing screenshots of code, database schemas, and diagrams. Also strong for text.

OpenRouter acts as a gateway to multiple providers through a single API key. Route specific models through preferred backends (e.g., Llama via Cerebras for speed).

Aura supports multiple API keys per provider via the apiKeys array in ai_providers.json:

"apiKeys": ["KEY_1", "KEY_2", "KEY_3"]

Why this is a game-changer:

A single key works perfectly fine — multiple keys are optional but recommended for heavy use.

🏆 Recommended setup: 2–3 free keys from Cerebras (primary, fastest) + 2–3 from Groq (secondary + vision) + 1 from Gemini (vision backup). Total cost: $0. Reliability: bulletproof.

The app auto-creates .env from .env.example on first run.

# ─── API Keys ───
DEEPGRAM_API_KEY="your_key"              # Required — Deepgram speech-to-text

# ─── Logging ───
LOG_LEVEL=INFO                            # DEBUG, INFO, WARNING, ERROR

# ─── Development ───
DEV_MODE=false                            # Verbose logging & dev shortcuts

# ─── AI Behaviour ───
TRACK_CANDIDATE_RESPONSES=true            # Track what the candidate says
INCLUDE_CONVERSATION_HISTORY=true         # Send conversation history to AI
MAX_CONVERSATION_HISTORY=6                # Past exchanges to include (6–20)
GENERATE_FULL_ANSWERS=true                # Full answers vs. brief hints
PERSONALIZE_ANSWERS=true                  # Tailor to your resume/JD

# ─── Stealth / Proctoring ───
ENABLE_SYSTEM_TRAY=false
START_IN_STEALTH_MODE=true

# ─── Scroll Speed (Alt+Up/Down) ───
SCROLL_SPEED_PX=200                       # Pixels per tick
SCROLL_INTERVAL_MS=50                     # Ms between ticks

Fine-tune how scrolling feels during a live interview:

Created from ai_providers.example.json. Not committed to git — your keys stay private.

{
  "name": "ProviderName",
  "baseURL": "https://api.provider.com/openai/v1",
  "apiKey": "YOUR_KEY",                    // Single key (minimum)
  "apiKeys": ["KEY_1", "KEY_2", "KEY_3"],  // Multiple keys for rotation
  "models": ["model-a", "model-b"],
  "visionModels": ["vision-model-a"],
  "supportsVision": true,
  "defaultModel": "model-a"
}

• Junior Developers — Compete with senior-level confidence with discreet access to best practices
• Career Changers — Navigate technical interviews in new domains with stealthy guidance
• Experienced Engineers — Excel in high-pressure FAANG interviews with an invisible partner for system design and algorithms
• International Candidates — Overcome language barriers with real-time phrasing suggestions, delivered discreetly
• Students — Ace aptitude tests and certifications using Vision AI in Stealth Mode

• Quantitative Analysts — Master complex math with on-screen, invisible formula assistance
• Professional Certifications — Navigate difficult cert exams with key information and strategies, undetectable
• Aptitude Tests — Crush pre-employment assessments by leveraging Vision AI to analyze diverse question types

Free. Fast. Flawless. And Fully Stealthy.

SCROLL_SPEED_PX=150      # Lower = slower
SCROLL_INTERVAL_MS=70    # Higher = less frequent

• No recurring fees or subscriptions for any feature, including advanced Stealth Mode
• Full access to all features — no premium tiers, no paywalls
• Your data stays on your machine — zero telemetry, zero tracking

• OpenAI-compatible — works with any provider using the standard API format
• Multiple providers — mix and match Cerebras, Groq, Gemini, OpenRouter
• Cost control — pay only for what you use, directly to providers (or use free tiers forever)
• No middleman — direct API access at provider rates

Aura's Stealth Mode is a powerful tool. It's designed to level the playing field and provide support—not to encourage dishonesty or misrepresentation.

BY USING AURA, INCLUDING ITS STEALTH MODE FEATURES, YOU ACKNOWLEDGE AND AGREE:

• Full Responsibility: Users are solely responsible for compliance with all applicable laws, regulations, company policies, and exam guidelines when using Stealth Mode
• Policy Review Required: Review your target company's interview policies or exam rules before using Aura
• Legal Compliance: Comply with all local, state, federal, and international laws regarding privacy, recording, and technology use
• Professional Ethics: Stealth Mode augments your abilities and confidence — it shouldn't misrepresent your fundamental skills

• Policy Violations: Using Stealth Mode in violation of policies may result in disqualification, offer withdrawal, or legal consequences
• No Guarantees: Aura does not guarantee success, job offers, or career advancement
• Technical Reliability: Issues may occur — don't solely rely on Aura without backup plans
• AI Limitations: Validate AI responses — don't blindly rely on them

• Enhancement, Not Deception: Use Aura to enhance your knowledge and confidence — not to fake competence
• Learning Tool: Treat Aura as an advanced study and performance aid
• Honest Representation: Maintain honesty about your capabilities
• Respect the Process: Respect the integrity of interviews and exams

By downloading, installing, or using Aura, you acknowledge that you have read, understood, and agree to comply with all terms and guidelines above.

git clone https://github.com/Rkcr7/Aura-AI
cd Aura-AI
Click run.bat to auto install dependecies
OR
python -m venv venv && venv\Scripts\activate
pip install -r requirements.txt
cp .env.example .env && cp ai_providers.example.json ai_providers.json
# Add your API keys, then:
python main.py

• 🐛 Bugs — Open an issue with steps to reproduce + terminal output
• 💡 Features — Describe the use case and expected behavior
• 🔧 PRs — Fork → branch → commit → pull request

This project is licensed under the MIT License — see LICENSE for details.

🏆 Aura: Victory Is Yours

Grab Aura. Crush it. Today.

⬆ Back to Top