Minecraft Loader Environment
CodeBreakers Logo
CODEBREAKERS PRESENTS
HACKVERSE '26 - Flagship Hackathon
> GENERATING TERRAIN CHUNKS...
0%

TIP: 24-HOUR NON-STOP BUILD SPRINT AHEAD

LIVE UPDATES
[WINNERS DECLARED] 🥇 JOINT 1ST: Team FPS (GCEK) & Team BRUTEFORCE (GITAM) • 🥈 2ND: CODE WARRIORS (GCEK) • 🥉 3RD: Team ACEBits (GCE Keonjhar) • 🌸 BEST FEMALE-LED: code_crash (GCEK) — Check out the full podium at /winners[THANK YOU] HACKVERSE '26 HAS OFFICIALLY CONCLUDED — Heartfelt thanks to all 500+ hackers, 60+ squads & mentors![ANNOUNCEMENT] HACKVERSE 2.0 COMING SOON — The saga continues! Bigger arena, larger prize pools & nationwide tracks[STAY TUNED] Connect with CodeBreakers GCEK on Instagram (@gcek.codebreakers) & LinkedIn for HackVerse 2.0 updates[WINNERS DECLARED] 🥇 JOINT 1ST: Team FPS (GCEK) & Team BRUTEFORCE (GITAM) • 🥈 2ND: CODE WARRIORS (GCEK) • 🥉 3RD: Team ACEBits (GCE Keonjhar) • 🌸 BEST FEMALE-LED: code_crash (GCEK) — Check out the full podium at /winners[THANK YOU] HACKVERSE '26 HAS OFFICIALLY CONCLUDED — Heartfelt thanks to all 500+ hackers, 60+ squads & mentors![ANNOUNCEMENT] HACKVERSE 2.0 COMING SOON — The saga continues! Bigger arena, larger prize pools & nationwide tracks[STAY TUNED] Connect with CodeBreakers GCEK on Instagram (@gcek.codebreakers) & LinkedIn for HackVerse 2.0 updates
HackVerse Problem Statements Minecraft Background
OFFICIAL PROBLEM STATEMENTS
6 LIVE CHALLENGES
HARDWARE & IOT AUTOMATION
HEALTHCARE, DISASTER & SUSTAINABILITY
OFFLINE DIGITAL LEARNING & AGRITECH
₹35K+ PRIZE POOL
OFFICIAL PROBLEM STATEMENTS
6 LIVE CHALLENGES
HARDWARE & IOT AUTOMATION
HEALTHCARE, DISASTER & SUSTAINABILITY
OFFLINE DIGITAL LEARNING & AGRITECH
₹35K+ PRIZE POOL
OFFICIAL PROBLEM STATEMENTS
6 LIVE CHALLENGES
HARDWARE & IOT AUTOMATION
HEALTHCARE, DISASTER & SUSTAINABILITY
OFFLINE DIGITAL LEARNING & AGRITECH
₹35K+ PRIZE POOL
OFFICIAL PROBLEM STATEMENTS
6 LIVE CHALLENGES
HARDWARE & IOT AUTOMATION
HEALTHCARE, DISASTER & SUSTAINABILITY
OFFLINE DIGITAL LEARNING & AGRITECH
₹35K+ PRIZE POOL
CHALLENGES // 2026 ARENA

OFFICIAL PROBLEM STATEMENTS

Select your arena of technical combat. Each challenge represents real-world engineering constraints with concrete evaluation criteria and industry relevance.

FILTER:
STATUS:SHOWING 11 OF 11 TRACKS
CB-HW-01
HardwareAdvanced
ORG //DEPT. OF EE, GCE KALAHANDI
SQUADS SELECTED:
00 SQUADS
GreenTech & Sustainability

AI-POWERED SMART ELECTRICITY MANAGEMENT SYSTEM FOR CAMPUSES

Colleges and organisations run hundreds of fans, tube-lights, ACs, and other electrical loads that are frequently left running in empty rooms, driving up electricity bills and wearing out equipment faster than necessary. This problem calls for an AI-powered, occupancy-aware system that automatically manages electrical loads and gives facility teams real-time visibility into where energy is actually being wasted.

EXPECTED SOLUTION:
  • â–¸1. Sensing Layer • Occupancy detection — PIR motion sensors and/or ultrasonic sensors per room to detect genuine human presence (not just movement of curtains/fans), with dual-sensor logic to reduce false positives/negatives. • Current/energy sensing — non-invasive current sensors (e.g., clamp-on CT sensors) on key circuits to measure real-time power draw per room or per appliance group. • Ambient sensing (optional) — light and temperature sensors to support smarter decisions (e.g., don't turn on lights if daylight is sufficient; don't run AC at full blast if the room is already cool).
  • â–¸2. Automated Control Layer • Relay-based switching — microcontroller-driven (e.g., ESP32/Arduino-class boards) relay modules that can automatically switch fans, lights, and other loads on/off based on occupancy and rules. • Grace-period logic — a configurable delay before auto-switch-off (e.g., 5–10 minutes of no detected occupancy) to avoid abrupt cut-offs while someone is still present but momentarily still. • Manual override — a physical switch or app-based override so users can retain manual control when genuinely needed, without fighting the automation. • AC-specific control — smart IR/relay control for ACs that adjusts based on occupancy and, where possible, temperature setpoint optimisation.
CB-HW-02
HardwareIntermediate
ORG //DEPT. OF ECE, GCE Kalahandi
SQUADS SELECTED:
00 SQUADS
Smart Cities & Digital Governance

IR-BASED TEMPLE CROWD COUNTER & QUEUE MANAGEMENT SYSTEM

Major temples like Puri Jagannath or Konark see huge, unpredictable crowd surges, especially during festivals, but have no simple way to know current footfall or queue length until the crowd is already unmanageable. This problem calls for a straightforward IR-sensor-based system that counts pilgrims in real time and helps manage queues before they become a safety risk.

EXPECTED SOLUTION:
  • â–¸IR break-beam/reflective sensor pairs — installed at entry and exit points (and optionally at 1–2 intermediate queue checkpoints) to count people crossing in each direction.
  • â–¸Microcontroller-based counter unit — (e.g., Arduino/ESP32-class board) reads sensor pulses, calculates net occupancy (entries minus exits), and handles basic debouncing logic to avoid miscounts from people walking closely together.
CB-SW-01
SoftwareAdvanced
ORG //Venus Global Technology Pvt. Ltd.
SQUADS SELECTED:
00 SQUADS
AI, Data & Emerging Tech (EdTech / HR-Tech)

AI-DRIVEN CAREER INTELLIGENCE & EMPLOYMENT MANAGEMENT PLATFORM

Students, fresh graduates, and job seekers struggle to discover relevant opportunities, understand companies, prepare for interviews, and identify the skills needed for their target roles — while existing job portals only list vacancies and leave everything else to the candidate. This calls for a centralized, AI-driven Career Intelligence Platform that unifies job discovery, company research, interview preparation, skill-gap analysis, and application tracking into one intelligent ecosystem, while giving institutions a real-time view of student employability.

EXPECTED SOLUTION:
  • â–¸1. AI-Powered Job Discovery & Matching • Semantic job matching — uses embedding-based similarity (not keyword search) between parsed candidate resume/profile and multi-source job postings, ranked by true role fit. • Aggregated multi-source discovery — pulls and de-duplicates postings from job boards and company career pages into one unified, refreshed feed. • Explainable match scoring — shows exact fit reasoning (overlap in specific skills, missing experience gaps) rather than a black-box score. • Fit-vs-stretch classification — automatically tags roles as 'safe fit', 'stretch', or 'reach' based on the gap between required and current skills.
  • â–¸2. Company & Employer Intelligence • AI-synthesized company profiles — culture, hiring process stages, typical interview formats, and compensation signals generated by summarizing public sources with citations. • Clear provenance labeling — every data point tagged as verified employer-provided, aggregated from public reviews, or AI-inferred. • Interview-experience mining — NLP extraction of recurring themes (question types, difficulty, round structure) from publicly shared interview experiences.
CB-SW-02
SoftwareAdvanced
ORG //Venus Global Technology Pvt. Ltd.
SQUADS SELECTED:
00 SQUADS
AI, Data & Emerging Tech (Smart Manufacturing / Industry 4.0)

AI-POWERED SMART MANUFACTURING ERP FOR INTELLIGENT PRODUCTION, AUTOMATION & BUSINESS OPERATIONS

Manufacturing businesses typically run production, inventory, procurement, workforce, quality control, and finance through disconnected systems and manual processes, leaving management with no real-time, unified view of operations. This problem calls for a centralized, AI-driven Smart Manufacturing ERP that connects every core business function on one platform while using AI to predict failures, optimize production, and automate routine decisions.

EXPECTED SOLUTION:
  • â–¸1. Production Planning & Scheduling • AI-optimized scheduling — generates production schedules based on demand forecasts, machine capacity, and resource availability, and re-optimizes dynamically when urgent orders or disruptions occur. • What-if simulation — lets planners simulate the impact of a new order or a resource change before committing to a schedule change.
  • â–¸2. Inventory & Supply Chain Management • Real-time stock tracking — live visibility into raw materials, work-in-progress, and finished-goods inventory across locations. • Demand-aware reorder recommendations — AI-driven suggestions for reorder points and quantities that reduce both overstocking and stockouts. • Procurement integration — connects inventory shortfalls directly to procurement workflows for faster replenishment.
CB-SW-03
SoftwareIntermediate
ORG //DEPT. OF CSE, GCE Kalahandi
SQUADS SELECTED:
00 SQUADS
EduTech & Future Learning

OFFLINE-FIRST DIGITAL LEARNING APP FOR UNDER-RESOURCED SCHOOLS

Many schools in rural and low-connectivity areas lack updated computer infrastructure, reliable internet, and access to quality digital educational content, widening the gap between rural and urban education standards. This problem calls for a mobile and web learning app that works genuinely offline and in local languages.

EXPECTED SOLUTION:
  • â–¸Offline-first mobile and web app — core lessons and content accessible without an active internet connection.
  • â–¸Interactive lessons in local languages to improve engagement and comprehension.
CB-SW-04
SoftwareAdvanced
SQUADS SELECTED:
00 SQUADS
AgriTech & Rural Innovation

UNIFIED AI FARM COPILOT, RISK INTELLIGENCE & MARKET OPTIMIZER

Farmers today need three different kinds of support that are rarely offered together: day-to-day crop guidance, early warning against risks that could wipe out a harvest, and smart decisions about what to grow and where to sell it. This problem calls for a single AI copilot that covers the entire journey — from sowing advice, through risk monitoring, to the final sell decision — instead of forcing farmers to piece together three different tools.

EXPECTED SOLUTION:
  • â–¸Module 1 — Farm Copilot (Personalised Advisory) • Multilingual voice & text interface — farmers ask questions or report issues by speaking naturally in their own language/dialect. • Image-based crop analysis — farmers upload/photograph crop or leaf images for AI-based identification of visible issues (disease symptoms, nutrient deficiency, pest damage). • Weather-aware guidance — combines local weather forecasts with crop stage to time irrigation, spraying, and other operations correctly. • Farm-history-aware personalisation — recommendations account for the specific farmer's past crop choices, soil type, and recorded outcomes, not generic advice. • Full crop-cycle coverage — guidance spans sowing, growth-stage care, harvest timing, and post-harvest handling, not just a single moment in the season.
  • â–¸Module 2 — Crop Risk Intelligence (Early Warning) • Multi-threat risk modelling — predicts risk across disease outbreaks, pest infestations, extreme weather events, water stress, and market-price volatility, rather than covering only one threat type. • Explainable alerts — every warning states why the risk is elevated (e.g., "high humidity + recent rainfall + susceptible crop stage → elevated fungal-disease risk") rather than issuing an opaque red flag. • Severity & urgency ranking — helps farmers prioritise when facing multiple simultaneous risks. • Mitigation guidance — each alert comes paired with concrete, actionable mitigation steps, not just a warning. • Feeds back into the Copilot — risk alerts surface directly within the same conversational interface farmers already use for daily advisory, rather than a separate app.
CB-SW-05
SoftwareIntermediate
ORG //ECO CLUB, GCE Kalahandi
SQUADS SELECTED:
00 SQUADS
GreenTech & Sustainability

GREENSCORE — CAMPUS SUSTAINABILITY TRACKER & AUDITOR

Colleges want to track and reduce their environmental footprint — energy, water, waste, and transport — but have no structured tool to actually measure it, motivate improvement, or produce the reports accreditation bodies expect. This problem calls for a campus sustainability app that captures utility data, scores performance, gamifies reduction efforts, and auto-generates audit-ready reports.

EXPECTED SOLUTION:
  • â–¸1. Data Capture Layer • Manual logging — simple mobile/web forms for students/staff/facility teams to log utility readings (electricity meter readings, water usage, waste weight/volume by category, transport mode for commutes). • Sensor-based auto-capture (where available) — integration with smart meters or IoT sensors (energy meters, water flow meters, smart bins) to pull readings automatically, reducing manual effort and improving accuracy. • Hybrid input model — works seamlessly for colleges with zero sensors (fully manual) as well as those with partial or full IoT infrastructure, without forcing hardware investment as a prerequisite. • Department/hostel-level granularity — data tagged by building, department, or hostel block to identify hotspots of high consumption.
  • â–¸2. Sustainability Scoring Engine • Monthly sustainability score — composite score computed from energy, water, waste, and transport metrics, normalised against campus size/population for fair cross-scale evaluation. • Category-wise breakdown — score decomposes into sub-scores per category to pinpoint strengths and areas needing improvement. • Trend tracking — month-over-month and year-over-year comparison with clear visual trend lines. • Benchmarking — comparison against historical campus baseline and/or anonymised comparison against similar-sized institutions.
CB-SW-06
SoftwareAdvanced
SQUADS SELECTED:
00 SQUADS
HealthTech & Wellbeing

AI-POWERED GAMIFIED FITNESS & LOCALIZED NUTRITION COPILOT

Most fitness apps suffer from high drop-off rates because they assume expensive wearables, push unrelatable Western diets, and offer no real local community engagement. This problem calls for an AI-driven, zero-hardware fitness platform tailored to the Indian and specifically Odia context — using just a smartphone camera for exercise tracking and building around local dietary staples and community-driven motivation.

EXPECTED SOLUTION:
  • â–¸1. Zero-Hardware AI Tracking (Computer Vision) • Real-time pose estimation — uses the smartphone's front camera to track joint movements during home workouts (squats, push-ups, yoga asanas), giving real-time audio-visual feedback on posture and automatically counting reps. • Low-end device optimisation — AI models run efficiently on edge devices, without requiring high-speed internet or a flagship processor. • Injury-risk flagging — detects clearly unsafe form (e.g., dangerous knee/back alignment) and proactively warns the user before it becomes a repeated habit, rather than only correcting rep count.
  • â–¸2. Localized Nutrition Engine • Regional diet mapping — a diet recommendation and tracking module built on recognised Indian nutritional guidelines, featuring local Odia cuisine (Mandia/millet-based dishes, Dalma, Pakhala) broken down by macronutrients. • Affordable alternatives — suggests low-cost, locally sourced protein and vitamin alternatives based on the user's budget and geographic location, rather than generic "eat more avocado" advice. • Meal-photo logging — lets users log meals by photographing their plate, with AI-assisted recognition of common local dishes to reduce manual entry friction.
CB-SW-07
SoftwareAdvanced
ORG //ADP Green Energy Pvt. Ltd
SQUADS SELECTED:
00 SQUADS
CleanTech & Renewable Energy

AI-POWERED SOLAR UNDERPERFORMANCE DIAGNOSIS & ROOT-CAUSE ENGINE

Solar photovoltaic (PV) power plants routinely experience drops in expected energy generation, forcing operations and maintenance (O&M) teams to spend critical hours manually identifying the root cause. This problem calls for an intelligent, telemetry-driven diagnostic engine that continuously compares expected versus actual generation, isolates the underlying failure mode, and delivers direct, prescriptive recommendations to maximize energy yield and asset life.

EXPECTED SOLUTION:
  • â–¸1. Data Ingestion & Baseline Modeling Layer • SCADA & telemetry ingestion — continuous ingestion of high-resolution time-series data from inverters (AC active/reactive power, DC voltage, DC current per MPPT or string, internal heat-sink temperatures), weather monitoring stations (Plane of Array [POA] irradiance, Global Horizontal Irradiance [GHI], ambient temperature, module surface temperature, wind speed), and grid feed-in status. • Physics-informed expected-generation modeling — high-fidelity digital-twin modeling (e.g., PVlib, Sandia/CEC inverter models) that computes dynamic theoretical power output (Pexp) in real time based on observed POA irradiance, module temperature coefficients, and system design specifications. • Data sanity & sensor validation — automated data-cleansing routines to detect pyranometer calibration drift, stuck values, and telemetry dropouts before passing inputs to downstream diagnostic models.
  • â–¸2. Root-Cause Diagnostic Engine • Yield-gap quantification — continuous evaluation of the delta between expected and actual generation (ΔP = Pexp − Pact), with dynamic filtering for transient cloud passage and rapid irradiance shifts. • Multi-class fault fingerprinting — rule-guided inference and machine-learning classifiers trained to distinguish between specific operational failure modes: uniform soiling vs. partial shading, inverter derating & clipping, string mismatch & open circuits, grid curtailment & external limits, and sensor calibration drift & pyranometer errors.
CB-SW-08
SoftwareIntermediate
SQUADS SELECTED:
00 SQUADS
HealthTech & Wellbeing

TELEHEALTH BRIDGE FOR UNDERSERVED RURAL AREAS

Small-town and rural hospitals often run at less than half their required doctor strength, forcing patients into long, often futile trips for care. This problem calls for a multilingual telemedicine solution built for genuinely low-connectivity rural conditions.

EXPECTED SOLUTION:
  • â–¸Multilingual video consultation app connecting rural patients directly with available doctors/specialists.
  • â–¸Offline-accessible digital health records so patient history is available even without a live connection.
CB-SW-09
SoftwareAdvanced
SQUADS SELECTED:
00 SQUADS
Smart Cities & Digital Governance

INTELLIGENT MULTI-HAZARD DISASTER MANAGEMENT & RESPONSE SYSTEM

Coastal and flood-prone regions face multiple, overlapping hazards — cyclones, floods, droughts, heatwaves, lightning, landslides, and storm surges — but early-warning systems usually stop at 'a warning was issued,' not 'here is what will happen where, and what to do about it.' This problem calls for an AI-powered, geospatial platform that turns real-time hazard data into location-specific risk assessments and actionable response decisions across multiple disaster types at once.

EXPECTED SOLUTION:
  • â–¸Hazard Detection & Ingestion Layer — pulls in real-time and historical data from weather feeds, satellite/open data, river-level sensors, and crowdsourced citizen reports, normalising heterogeneous sources into a common data model.
  • â–¸Risk Assessment Engine — combines hazard, exposure, and vulnerability data to compute location-specific risk, not just raw hazard intensity.
HACKVERSE '26
24-HOUR SPRINT
₹35K+ PRIZE POOL
CODEBREAKERS GCEK
AI / ML TRACK
CYBERSECURITY ARENA
BHAWANIPATNA ODISHA
HACKVERSE '26
24-HOUR SPRINT
₹35K+ PRIZE POOL
CODEBREAKERS GCEK
AI / ML TRACK
CYBERSECURITY ARENA
BHAWANIPATNA ODISHA
HACKVERSE '26
24-HOUR SPRINT
₹35K+ PRIZE POOL
CODEBREAKERS GCEK
AI / ML TRACK
CYBERSECURITY ARENA
BHAWANIPATNA ODISHA
HACKVERSE '26
24-HOUR SPRINT
₹35K+ PRIZE POOL
CODEBREAKERS GCEK
AI / ML TRACK
CYBERSECURITY ARENA
BHAWANIPATNA ODISHA