Skip to content
ibochivincent-langPublic

About

Resources

Stars

1 star

Watchers

0 watching

Forks

Latest commit

 

History

205 Commits

Folders and files

NameName
Last commit message
Last commit date
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Career Ace: Universal Autonomous AI Career Copilot & Decentralized Vault

Built with Gemini Live Demo Walrus Mainnet Certified Sui Network Next.js 16 Tests Passing Memory Evaluation Zero Mock Data

Author: IboTV (ibochivincent-lang) — Sole Author and Maintainer.
AI Engine: Built with Google Gemini (@google/genai & @ai-sdk/google).
Production URL: https://careerace.online
Repository: https://github.com/ibochivincent-lang/careerace
Stack: Next.js 16 (Turbopack), React 19, TypeScript 5 (Strict), Tailwind CSS v4, Google Gemini AI Engine, Walrus Protocol, Sui Blockchain, Mysten Labs SDK, Google zkLogin, SuiNS.

Career Ace: Autonomous AI Career Copilot & Decentralized Vault


Hackathon Submission Overview

Career Ace is an autonomous AI career copilot, dual-track application automator, multi-industry CV tailor, decentralized career vault, and STAR+R interview coach.

Traditional career platforms suffer from three structural flaws:

  1. Platform Data Lock-In: Centralized portals hold candidate milestone data hostage, commoditize candidate resumes, and monetize access to proprietary profile databases.
  2. Opaque Applicant Tracking System (ATS) Scoring: Enterprise parsers (Workday, Taleo, Greenhouse, Lever) reject over 75% of qualified technical candidates due to minor typographical or single-column layout formatting mismatches.
  3. Generic AI Hallucinations: Standard LLM resume tools invent fake credentials, generate buzzword-bloated phrasing, and expose sensitive candidate records to third-party model retraining.

Career Ace solves this permanently:

  • Candidate career milestones, credentials, and tailored resumes are encrypted via AES-256-GCM and stored across decentralized Walrus Protocol storage nodes.
  • Cryptographic anchors and human-readable .sui passports are registered directly on the Sui blockchain via Sui Move smart contracts and Sui Name Service (SuiNS).
  • Candidates access a private, autonomous cockpit with reverse-engineered ATS scoring, zero mock data hiring directories (382+ verified employers), multi-provider authenticated email dispatch, and multi-turn interview coaching with refusal guards.

Targeted Hackathon Tracks

Track Implementation & Verified Deliverables
Walrus Protocol Decentralized Storage 15 Certified Blobs on Walrus Mainnet (WALRUS_MAINNET_PROOF.md), client-side AES-256-GCM encrypted profile storage, content-addressed blob retrieval across epochs, versioned snapshot drawer, and persistent candidate memory.
Sui Network & zkLogin Identity Frictionless Web2 onboarding via Google zkLogin (zero seed-phrase friction), Sui Move smart contracts for onchain cryptographic proof anchors, and human-readable .sui domain resolution for public passports (/p/[username]).
Autonomous AI Agents & ATS Intelligence Sub-350ms Walrus memory recall copilot, 200 intent-categorized candidate question taxonomy, Google XYZ formula ATS resume audit, and multi-turn STAR+R pedagogical interview simulator.

Verified Walrus Mainnet Certification

Career Ace has written and certified 15 production memory blobs on Walrus Mainnet under Account ID 0x434f860c828dc4320be447975b8283d7c5786c4a08b9ddc8f88540d9ea69aa00, exceeding hackathon criteria:

# Candidate Namespace Category Blob ID Walrus Mainnet Aggregator Proof
1 Alex Rivera candidate_alex_rivera Target Role wjP72rBhrofqKKnXb09zhFHIQV5cn2m8pBNg7PP89kw Inspect on Mainnet
2 Alex Rivera candidate_alex_rivera Work Experience v2TnnTVm5GSHfuUUgxNv_Xv765uJV9T-XjLMURGyB6A Inspect on Mainnet
3 Alex Rivera candidate_alex_rivera Key Metric DeRSqFzXUdpXNcJ1HLUm8RFG8in0OFxRBgbJuEE2XmY Inspect on Mainnet
4 Alex Rivera candidate_alex_rivera Core Competency mOnV2i6nxlyVv5TrQhCTzAVyoYWgdivaUS0gM_VTFy8 Inspect on Mainnet
5 Alex Rivera candidate_alex_rivera Interview Strategy _OSyUNcyWODBdkcpXGGOazobl3LxF0YS8DmKfh5fheo Inspect on Mainnet
6 Sarah Chen candidate_sarah_chen Target Role pvEU6hNfe7kkLdR6jUO4a84oH5exQLb_dyCeliksi0E Inspect on Mainnet
7 Sarah Chen candidate_sarah_chen Work Experience rROuuDYYBJVs1ExX7ib2IYHJCE2TK9zTrRZpC3k7wD0 Inspect on Mainnet
8 Sarah Chen candidate_sarah_chen Key Metric 8UNPnFcoxgq9Nbg1E7ocm9ZZ2kS1LOS3I3gyKioMb4g Inspect on Mainnet
9 Sarah Chen candidate_sarah_chen Core Competency m3LOcgj7D7VbdZOE9P1xr-fmjXCSnZXIT6iLfPzer_c Inspect on Mainnet
10 Sarah Chen candidate_sarah_chen CV Bullet lmih4_JEvtM23CxzodqcQIGmTGjcdAOkRolcHvvkYBI Inspect on Mainnet
11 Marcus Adebayo candidate_marcus_adebayo Target Role B6gBL5Tk2S50SHOJ8QgzA-HMByITC3MnxPEcHd5A0LY Inspect on Mainnet
12 Marcus Adebayo candidate_marcus_adebayo Work Experience flWH9IuiwldSs3DBkfC9HXJRPbw_5oX8aZf6oLNQjZs Inspect on Mainnet
13 Marcus Adebayo candidate_marcus_adebayo Key Metric kSkUnxcBVMv12wv5TaSZW9kz7wKC6JHKdZL1wwJo8gc Inspect on Mainnet
14 Marcus Adebayo candidate_marcus_adebayo Core Competency VFUNOVesikNLrH7ZUvOV8hFo1CuvRqfRhn-_ExyTXTE Inspect on Mainnet
15 Marcus Adebayo candidate_marcus_adebayo CV Bullet nbGHckr34yB4BQGiXYsz9wy8AC2zAlX4zMuqic1iGdc Inspect on Mainnet

For verification logs and semantic recall benchmarks, see WALRUS_MAINNET_PROOF.md.


Master Architecture Flow

sequenceDiagram
    autonumber
    actor Candidate as Candidate / Engineer
    participant Web as Next.js 16 Client (React 19)
    participant API as Next.js API Routes (Edge Engine)
    participant Copilot as Career Copilot & ATS Parser
    participant Walrus as Walrus Protocol Storage Nodes
    participant Sui as Sui Blockchain (Move Contract & SuiNS)
    participant Email as Multi-Provider Relay (Resend/Brevo/SendGrid)
    participant Employer as Verified Employer Inbox

    Candidate->>Web: 1. Sign in via Google zkLogin
    Web->>Sui: Derives sovereign address (0x...) with zero seed-phrase friction
    Candidate->>Web: 2. Uploads CV (PDF, DOCX) & documents
    Web->>API: POST /api/cv_upload
    API->>Copilot: Heuristic multi-section extraction (Google XYZ metrics)
    Copilot-->>API: Parsed profile & ATS score (Workday/Taleo/Greenhouse)
    API->>Walrus: Store encrypted profile & raw documents (AES-256-GCM)
    Walrus-->>API: Returns content-addressed Blob ID
    API->>Sui: Register onchain anchor & bind candidate.sui handle
    Sui-->>Web: Confirmed onchain transaction digest & passport URL (/p/[username])

    Candidate->>Web: 3. Interacts with Copilot & Two-Box Dispatch Board
    Web->>Copilot: Asks career question / selects curated target job
    Copilot->>Walrus: Semantic memory query (<350ms recall)
    Walrus-->>Copilot: Recalls historical verified milestones & metrics
    Copilot-->>Web: Delivers grounded edit / tailored pitch without AI slop

    Candidate->>Web: 4. Dispatches application with Walrus attachments
    Web->>API: POST /api/email/dispatch
    API->>Email: Cascades primary Resend -> Brevo -> SendGrid with DKIM/SPF
    Email-->>Employer: Delivers signed email with Walrus-hosted attachments & RFC 5545 calendar reminder
    Email-->>Web: Verified dispatch report with RFC 5322 EML confirmation
Loading

Universal Technical Domains & Industry Coverage

Career Ace operates across six core technical sectors with specialized vocabulary models, credential gates, and direct corporate recruitment directories:

Sector Target Roles Verification & Credential Standards Sample Employers (382+ Directory)
Software & Distributed Cloud Staff Backend Engineer, Cloud Architect, Site Reliability Engineer AWS / GCP / Azure Certifications, CKA (Kubernetes), CNCF, RFCs Stripe, AWS, Cloudflare, Datadog, Snowflake
AI, ML & Autonomous Robotics AI Research Scientist, ML Engineer, Robotics Systems Architect PyTorch, CUDA kernel optimizations, arXiv publications, ROS 2 Google DeepMind, OpenAI, Boston Dynamics, Anduril
Maritime & Offshore Systems (Flagship Case Study) Chief Marine Engineer, DP Officer, Vessel Superintendent, Naval Architect STCW Basic Safety (BST), CoC Class 1-4, ENG1 Medical, DP Maintenance, BOSIET Maersk, ABS, Stolt-Nielsen, Bourbonese, Tidewater, Subsea 7, DNV
Healthcare & Biomedical Tech Health Informatics Lead, Clinical Data Engineer, Biomedical AI Specialist FDA 21 CFR Part 11, HIPAA Compliance, HL7 / FHIR, GCP Clinical Epic Systems, Roche, Illumina, Mayo Clinic
Industrial & Electrical Systems SCADA / PLC Engineer, Grid Automation Specialist, Instrumentation Lead IEEE Standards, Siemens TIA, Rockwell ControlLogix, SIL 3 Safety Siemens, ABB, Schneider Electric, Rockwell Automation
Cybersecurity & Infrastructure Threat Hunter, Penetration Tester, SecOps Lead, Zero-Trust Architect CISSP, OSCP, SOC 2 Type II Auditing, NIST CSF, ISO 27001 CrowdStrike, Palo Alto Networks, Cloudflare, Okta

Flagship Case Study — Maritime & STCW Engineering: Maritime engineering represents the most stringent credential verification regime globally under International Maritime Organization (IMO) conventions. Candidates must prove physical licenses (CoC Class 1-4), safety training (BST, BOSIET), dynamic positioning qualifications, and medical fitness (ENG1). By designing deterministic parsers and verifiers capable of handling these physical credentials, Career Ace proves its architecture can enforce any strict compliance standard across all modern industries.


End-to-End Feature Tour

1. The Autonomous AI Career Agent (/dashboard)

  • Fast Walrus Sovereign Memory Recall: Memory lookups execute under 350ms bounded timeouts, seamlessly injecting candidate milestones into LLM prompts without lag.
  • 200 Simulated Questions Taxonomy: Instant prompt chips categorized across 6 key candidate domains (Skills & Stack, Experience & Metrics, Work Authorization, Career Goals, Strengths & Weaknesses, and Verification).
  • 2x2 Cockpit Grid: Balanced interface presenting Recommended Jobs, Application Tracker, Daily Limits & Goals, and Profile & Qualifications with direct version switching.

2. Reverse-Engineered ATS X-Ray (/dashboard?tab=overview)

  • Multi-ATS Simulation: Tested against parsing algorithms for Workday, Taleo, Greenhouse, and Lever.
  • Google XYZ Formula Enforcement: Identifies weak passive phrasing and rewrites bullets into quantified impact statements (Accomplished [X], as measured by [Y], by doing [Z]).
  • Single-Column Standards: Ensures zero-table, single-column DOCX and JSON Resume exports that never choke enterprise parsers.

3. Two-Box Auto-Apply Dispatch Console (/application_board)

  • Box 1 (Document Attachments & Walrus Blobs): Multi-file native upload (.pdf, .docx, .doc, .png, .jpg) cached in browser storage and permanently anchored as Walrus blobs with file size and status chips.
  • Box 2 (Inline Cover Letter & Pitch Editor): Full-width live editor with dynamic company greetings, user edit preservation (isBodyUserEdited), reset controls, and character counters.
  • 1-Click Authenticated Email Relay: Dispatches applications via authenticated SMTP/API with automatic failover across Resend, Brevo, and SendGrid, including RFC 5545 .ics follow-up calendar invites.

4. Sovereign Memory Vault (/memory)

  • Blob Explorer: Real-time inspection of encrypted Walrus blobs, transaction digests, and epoch allocations.
  • Snapshot Drawer (WalrusVersionDrawer): 1-click restore to past versions of candidate CVs without data loss.

5. STAR+R Pedagogical Interview Coach (/interview_room)

  • Socratic Simulation: Evaluates candidate responses across Situation, Task, Action, Result, and Reflection.
  • Pedagogical Refusal Guards: Refuses to hand out direct answers, pushing candidates to uncover their own architectural tradeoffs through guided questioning.

6. Public Verifiable Passports (/p/[username])

  • SuiNS Name Binding: Resolves .sui human-readable handles (e.g. alex.sui) and serves a verifiable public candidate passport backed by cryptographic onchain anchors.

Complete API Route Specifications

Endpoint Method Description Primary Engine / Storage
/api/cv_upload POST Ingests PDF/DOCX resumes, executes heuristic multi-section parsing, and extracts quantified metrics OpenXML / PDFParse / Heuristic Engine
/api/attachment/upload POST Stores multi-file application attachments directly to Walrus Protocol nodes Walrus Publisher REST API
/api/walrus/anchor POST Generates onchain Move cryptographic anchors linking candidate address to Walrus Blob ID Sui Move Smart Contract Layer
/api/memory GET/POST Stores and recalls candidate sovereign memory facts across isolated namespaces Walrus Memory Aggregator & Relayer
/api/suins/bind POST Normalizes and binds .sui human-readable handles to candidate sovereign addresses SuiNS Resolver Engine
/api/suins/resolve GET Resolves .sui handles into candidate sovereign passports and metadata Sui RPC / SuiNS Protocol
/api/ats/benchmark POST Simulates enterprise ATS scoring (Workday, Taleo, Lever, Greenhouse) Multi-ATS Simulation Algorithm
/api/tailor POST Aligns candidate CV bullets with specific job descriptions using Google XYZ metrics Grounded Tailoring Engine (No AI Slop)
/api/email/dispatch POST Dispatches DKIM/SPF authenticated application emails with multi-provider failover Resend / Brevo / SendGrid Relay
/api/harvest GET/POST Harvests live job postings from 9 real-time networks without mock data Live Job Crawlers (Jobicy, Remotive, Arbeitnow)
/api/interview POST Multi-turn STAR+R interview coach with pedagogical refusal guards Socratic Evaluation Controller
/api/feedback POST Ingests validated candidate feedback with rate-limiting and database persistence Supabase / Database Engine

Technology Stack

Frontend:           Next.js 16 (App Router, Turbopack), React 19, Tailwind CSS v4, Lucide Icons, Motion
Language:           TypeScript 5 (Strict Mode)
Decentralized Data: Walrus Protocol (Mysten Labs), Memwal Relayer
Blockchain:         Sui Network (Testnet & Mainnet), Sui Move, SuiNS, Enoki zkLogin
Document Engine:    PDFParse, Mammoth (DOCX), OpenXML Spec, JSON Resume Schema v1.0.0
Email Relay:        Resend API (multi-key rotation), Brevo API, SendGrid API, RFC 5322 EML, RFC 5545 ICS
Testing Suite:      Node.js 20+ Native Test Runner (Zero Bloat), 193 Tests Passing across 11 suites

Local Development & Setup

Prerequisites

  • Node.js 20.x or higher
  • pnpm (recommended) or npm
  • Git

Installation

# 1. Clone the repository
git clone https://github.com/ibochivincent-lang/careerace.git
cd careerace

# 2. Install dependencies
pnpm install

# 3. Configure environment variables
cp .env.example .env.local

# 4. Start the development server with Turbopack
pnpm run dev

Open http://localhost:3000 in your browser.


Testing & Verification Suite

Career Ace maintains an automated test suite verifying CV parsing, credential gating, SuiNS normalization, Walrus anchoring, ATS scoring, multi-provider email cascades, RFC 5322 EML generation, RFC 5545 calendar generation, and the 15-section Chatbot Memory Evaluation & QA Checklist (Tests A through J):

# Run full automated test suite (193 tests across 11 test suites)
npm test

# Run dedicated Chatbot Memory QA evaluation suite (Tests A-J)
node --test --experimental-strip-types lib/chatbot_memory_qa.test.ts

# Run TypeScript static type checking (0 errors)
npx tsc --noEmit

# Run Next.js production build verification (62 routes compiled)
npm run build

Detailed test outputs, memory boundary verification, and framework matrices are documented in CHATBOT_MEMORY_EVALUATION.md.


Deployment Guide: What & Where to Deploy

1. Web Application & API Routes: Vercel

  • Repository: Connected to ibochivincent-lang/careerace (main branch).
  • Framework Preset: Next.js
  • Build Command: npm run build
  • Environment Variables:
    • NEXT_PUBLIC_APP_URL=https://careerace.online
    • WALRUS_PUBLISHER_URL=https://publisher.walrus-mainnet.walrus.space
    • WALRUS_AGGREGATOR_URL=https://aggregator.walrus-mainnet.walrus.space
    • SUI_NETWORK=testnet
    • SUI_RPC_URL=https://fullnode.testnet.sui.io:443
    • RESEND_API_KEY (and optional RESEND_API_KEY_2, BREVO_API_KEY, SENDGRID_API_KEY)
    • GEMINI_API_KEY (or Google Generative AI key for live copilot inference)

2. Decentralized Storage: Walrus Protocol

  • Profile snapshots, document attachments, and candidate memory are written directly to Walrus storage nodes via HTTP publisher APIs. No dedicated backend server required.

3. Blockchain & Identity: Sui Network

  • Onchain anchors and SuiNS passport resolution execute against the Sui RPC (https://fullnode.testnet.sui.io:443).

World-Class Standard Roadmap

  1. Inbound Recruiter Webhook: Capture employer replies (/api/email/inbound) and automatically update Application Tracker statuses.
  2. Onchain Soulbound Credential Badges (Sui Kiosk / W3C VC): Enable universities, classification societies, and certification boards to mint verified credentials directly into candidate Walrus vaults.
  3. Biometric WebRTC Speech Practice: Upgrade the STAR+R Interview Coach with real-time speech prosody and filler-word detection (um, like) during verbal simulation rounds.

Author & AI Engine Attribution

  • AI Engine: Built with Google Gemini (@google/genai & @ai-sdk/google)
  • Sole Author & Maintainer: IboTV (ibochivincent-lang)
  • License: MIT License

About

Resources

Stars

1 star

Watchers

0 watching

Forks

Releases

Packages

Contributors

Languages