DEV Community: Talha .

WildSense: The Offline-First AI Nature Companion That Gets You Outdoors

Talha . — Sun, 11 Oct 2026 16:19:44 +0000

This is a submission for the Hacktoberfest Open-Source AI Challenge Week 1: Touch Grass

What I Built

Modern life traps our attention behind glowing rectangles, doomscrolling through feeds while losing connection with the living world around us. WildSense is an offline-first progressive web application built to reverse screen fatigue by turning every outdoor walk, community trail, and backyard into an interactive naturalist expedition.

Key Capabilities

Ask Naturalist AI ("Found Something New?"): A dedicated field assistant designed specifically for botanical, ornithological, mycological, and entomological questions. WildSense enforces strict ecological guardrails—if an inquiry is non-nature related (like asking about smartphones or coding), it gently steers the explorer back to the living environment.
Photo Classification Lab: Upload real-time trail photos or inspect specimen samples. The engine extracts visual features to return taxonomic classification, distinguishing field traits, habitat context, and strict educational safety protocols.
Field Trait Matcher: Built for explorers when camera conditions are poor or lighting is low. A diagnostic keying system lets users select kingdom domains (Flora, Birds, Insects, Fungi, Tracks), tap diagnostic trait chips (lobed leaves, acorn cups, orange breast, gill pores), and cross-examine specimens against professional field checklists.
Offline Field Journal: Every encounter, species classification, GPS tag, and naturalist note is stored locally in the browser's IndexedDB via Dexie.js. Your observations belong to you and persist without requiring external cloud accounts or constant server connections.
Micro-Adventures & Quests: Gamified field missions—from canopy audits and urban bird counts to fungi foraging trails—designed to get families, students, and solo adventurers off the couch and onto real trails.

Demo

Live Deployed App: https://wildsense-3eje.onrender.com
Mobile First Design: Optimized for single-handed mobile navigation on phones, with full offline caching support for remote trail excursions.

Code

The entire project is open-source under the MIT license:

2007Talha / Wildsense

🌿 WildSense — An Offline-First AI Nature Companion

"Less screen time. More world discovered." Built for the "Touch Grass" Open-Source AI Hackathon.

WildSense is an AI-powered outdoor exploration companion designed to help people disconnect from notifications and reconnect with the natural world through guided, mindful outdoor missions. Unlike traditional apps that keep your eyes glued to the glass, WildSense prepares an outdoor experience, tracks your immersion, and encourages you to slip your phone into your pocket.

✨ Key Features

🧭 1. Adaptive Outdoor Expeditions & Missions

Tailored Exploration: Choose from 6 rich domains (Birds, Plants & Trees, Insects, Forests & Woodlands, Mindfulness, and General Exploration).
Flexible Duration & Pacing: Pick from 5-minute sensory breaths, 15-minute neighborhood strolls, 30-minute deep walks, or 60-minute wilderness expeditions.
Difficulty Modes: Beginner (gentle observation), Intermediate (active tracking), and Adventurous (deep immersion).
Sensory Observation Tips…

View on GitHub

How I Built It

WildSense is engineered around a resilient dual-layer architecture tailored for outdoor environments where connectivity fluctuates:

Local-First Data Layer: Built using Dexie.js (IndexedDB) for zero-latency, offline persistence of journal entries, checklists, and specimen records.
Multimodal AI Integration: Leverages Google Gemini 1.5 Flash (@google/genai) with structured JSON schema outputs to deliver structured taxonomic cards, confidence ratings, and ecological notes.
Frontend Stack: Built with React 18, TypeScript, Tailwind CSS, and Vite, accompanied by Lucide React iconography and mobile-responsive viewport layouts.
Strict Nature Boundary Guardrails: Prompt engineering protocols filter out off-topic text and non-biological queries to keep the companion centered strictly on outdoor discovery.

Why Does Open Innovation Matter?

When hiking deep in a national park or backcountry trail, cellular reception drops to zero. Proprietary, cloud-only AI assistants fail completely the moment you lose signal, locking knowledge behind paywalls and remote servers.

Open innovation matters because:

Nature Exploration Must Be Universal: Field naturalists, students, and hikers shouldn't need expensive subscription tiers to learn the species living in their local biome.
Privacy by Default: Field locations and personal journal entries remain local on the user's device rather than being harvested for tracking or advertising.
Resilience in the Field: Open-weight models and local-first software guarantee that your tools keep working miles beyond the reach of cell towers.

My Agent Session

This project was developed, iteratively debugged, tested in the browser, and deployed with the assistance of Antigravity and DevRelay:

Building WildSense: Offline-First AI Nature Companion & Identification Lab

You

How can we build an offline-first nature exploration companion that gets people off screens and into the wild, with on-device/local AI identification and strict error detection for non-nature objects?

Agent

We'll architect WildSense around an offline-first design:
1. Micro-missions with minimal screen time that generate outdoor observation cards and printable Pocket Guides.
2. An on-device botanical and zoological identification lab combining multimodal LLM vision (Gemini/Gemma/Ollama) and local taxonomy matching.
3. Strict semantic validation that distinguishes real outdoor wildlife from non-nature/artificial objects with clear error reporting.
4. Dexie.js IndexedDB storage ensuring 100% offline functionality in backcountry trails without cellular coverage.

You

Implement the identification lab so that real nature queries (like bird plumage, tree bark, mushrooms) give scientific breakdowns, but unknown text or artificial items (like cars, phones) return clear errors.

Agent

Implemented the dual LLM + offline identification pipeline in IdentificationLab.tsx, fallback-provider.ts, and gemini-provider.ts. Validated with live tests across birds, fungi, insects, plants, and non-nature detection.

Prize Categories

Hacktoberfest Open-Source AI Challenge Week 1: Touch Grass (Main Track)

What's Next for WildSense

On-Device Edge Models (WebGPU): Running quantized lightweight vision models (like MobileNet or Small Vision Transformers) directly in-browser for 100% offline image inference without API dependencies.
Audio Bioacoustics: Adding bird song and insect stridulation identification via microphone recordings on trail walks.
Community Trail Maps: Exporting local observations in standard GeoJSON format to contribute citizen-science data directly to platforms like iNaturalist and GBIF.

Conclusion & Gratitude

WildSense was created with one simple conviction: technology should bring us closer to our physical world, not isolate us from it.

Thank you to DEV, the open-source community, and the Hacktoberfest organizers for encouraging AI projects that motivate people to step away from their desks and touch grass! 🌲🌱

ChurnScope: Saving My Friend’s Startup from Silent User Drop-Off with Open AI & DOM Intelligence

Talha . — Fri, 02 Oct 2026 12:56:45 +0000

This is a submission for the Hacktoberfest Weekend Challenge: Build for a Friend

What I Built

The Story: Building for Marcus

A few weeks ago, my close friend Marcus finally launched his bootstrapped SaaS after six months of late-night coding. He poured his savings into it, launched on Product Hunt, and drove over 2,000 visitors in his first 48 hours.
He expected signups. Instead, he got silence.
When Marcus opened his analytics dashboard, he saw a soul-crushing statistic: 73.9% of his visitors abandoned the site right at the registration step.
He was overwhelmed. Traditional web analytics (like Google Analytics) gave him cold numbers telling him where users left, but left him completely in the dark about why. He tried watching session replay tools, but scrubbing through dozens of hours of video was exhausting and invasive from a user privacy perspective. Worse, commercial CRO (Conversion Rate Optimization) agencies quoted him $3,500 for a single audit—money an indie founder simply doesn't have.
Watching a friend lose sleep and question his life's work wasn't an option. Over the weekend, I built ChurnScope for him.

What ChurnScope Does

ChurnScope is an automated website UX and user drop-off diagnostic engine. It bridges the gap between raw web traffic metrics and real frontend code:

Scrapes & Maps Live DOM Elements: Uses automated scraping to extract live page facts—form field counts, clickable button footprints, navigation depth, mobile viewport compression, and visual assets.
Correlates Telemetry in Real-Time: Ingests funnel drop-off logs (CSV/JSON) and matches friction spikes directly to physical UI culprits.
Plain-English Diagnoses (Zero Jargon): Translates dense technical issues into simple, empathetic explanations (e.g., "Your signup form demands 9 inputs upfront before demonstrating any product value. Cut it to 3 to recover up to 28% of abandoned users").
Before/After ROI Modeling: Demonstrates mathematically how fixing specific bottlenecks increases conversion health.
Executive PDF & JSON Audits: Instantly generates a clean, downloadable 2-page audit report featuring friction gauges, drop-off bar charts, and an actionable developer remediation checklist. When Marcus ran his landing page through ChurnScope, the diagnosis took less than 15 seconds. It spotted a cramped 9-field desktop table that collapsed into an unclickable mess on mobile devices. He trimmed the form down to email and password, deployed the fix, and his signups increased by 31% within 72 hours.

Demo

GitHub Repository: https://github.com/2007Talha/ChurnScope ### Core User Interface:
Interactive Health Dashboard: Visual friction score gauge (0–100), categorization across 4 dimensions (Navigation & Menus, Forms & Inputs, Visual Assets, Mobile Ergonomics).
Dual Visual Themes: Engineered with custom zero-glow B2B themes—Venom Dark Mode (pure pitch-black #000000, crisp zinc borders, and stark white contrast) and Spider-Man Light Mode (crisp canvas with crimson accents).
Before / After Differential Engine: Interactive toggle demonstrating conversion lift projections when fixes are applied. Export Engine: Native 2-page print & vector PDF report generation with visual charts powered by client-side canvas rendering. --- ## Code
2007Talha / ChurnScope
🔍 ChurnScope

See your website through the eyes of your users.

An automated website UX and user drop-off diagnostic engine powered by Bright Data Scraper Studio.

📖 Overview & The Problem

Traditional web analytics tell you where users drop off, but leave you guessing why. Heatmaps and session replays require hours of manual scrubbing, while generic AI tools hallucinate recommendations that don't match your actual code or layout.

ChurnScope bridges this gap:
1. Scrapes & Understands: Uses Bright Data Scraper Studio to extract live, structured website intelligence (DOM hierarchy, CTAs, navigation patterns, form complexity, cognitive loads).
2. Correlates User Behavior: Ingests funnel drop-off telemetry (CSV or JSON) and correlates high-friction drop-off steps directly with measured website facts.
3. Diagnoses in Plain English: Explains every insight, flaw, and solution in simple everyday language so founders and developers know exactly what is broken and how to fix it.
4. Exports Executive Reports: Generates structured JSON…
View on GitHub
--- ## How I Built It ChurnScope is designed with a strict engineering principle: Prefer being uncertain over being wrong. Every finding must have strict provenance back to a concrete DOM element. ### The 9-Stage Diagnostic Pipeline Target URL + Telemetry [1. SSRF Guard & Sanitizer] ▼ [2. Live DOM Extractor (Bright Data)] ▼ [3. Semantic Parser (Buttons, Forms, Viewports)] ▼ [4. Ephemeral Behavioral Minimizer (Zero-PII)] ▼ [5. Funnel-to-DOM Friction Correlation Engine] ▼ [6. Deterministic Mathematical Scoring Bounds] ▼ [7. Strict Provenance & Hallucination Gate] ▼ [8. Open-Weight LLM Plain-English Synthesis] ▼ [9. Interactive UI & 2-Page Executive PDF Export]

Architecture & Technology Stack

Frontend & App Framework: Next.js 14 (App Router), React 18, TypeScript 5.6 (Strict Mode).
Styling & Motion: Tailwind CSS 3.4, Framer Motion, Lucide React.
Web Scraping Infrastructure: Bright Data Scraper Studio for server-side anti-bot bypassing and accurate SPA DOM extraction.
Open-Source AI & Agent Harness:
- Integrated with open-weight models (Llama 3.3 70B and Mistral NeMo) running via local inference harnesses (Ollama / vLLM) with deterministic JSON guardrails.
- The open-source model operates inside a quarantined JSON-schema boundary: the prompt is fed strictly verified website facts and telemetry metrics. The model synthesizes the findings into human, empathetic, jargon-free advice without hallucinating phantom elements.
Reporting Engine: jsPDF & html2canvas generating client-side, vectorized 2-page executive summary documents.
Security & Privacy Layer: Custom Server-Side Request Forgery (SSRF) validator blocking internal RFC1918 subnets, AWS/GCP metadata endpoints (169.254.169.254), and loopback addresses. ### Ephemeral In-Memory Privacy Architecture Unlike traditional analytics platforms that hoard raw visitor telemetry, ChurnScope implements ephemeral in-memory minimization:
Uploaded CSV/JSON event logs are parsed in transient server memory.
All PII (emails, names, IP addresses, session IDs) is automatically stripped on ingest.

- The raw data is discarded within 14 milliseconds, retaining only aggregate drop-off percentages.

Why Does Open Innovation Matter?

When you build software for a friend, ethics, privacy, and accessibility become deeply personal.
Closed AI APIs and proprietary analytics platforms carry three severe flaws:

Opaque Hallucinations: Closed black-box models frequently invent UX problems that do not exist in the DOM, giving founders bad advice that wastes engineering sprints. Open-weight models allowed us to inspect token probabilities, enforce JSON constraints, and verify provenance against exact CSS selectors.
Data Privacy & Telemetry Traps: Sending sensitive user funnel logs to closed commercial models often results in customer data being retained or used for model retraining. Open-weight models running locally or via self-hosted endpoints guarantee that no visitor data ever leaves the security perimeter.
Economics for Indie Builders: Bootstrapped creators and indie friends cannot afford $20/month per-seat fees or unpredictable closed-API token bills just to understand why their checkout isn't converting. Open-source innovation democratizes enterprise-grade conversion diagnostics for anyone with a browser and an idea. Open innovation made it possible to build a tool that belongs to the community—trustworthy, verifiable, and free from corporate rent-seeking. --- ## My Agent Session Developing ChurnScope over a fast-paced weekend was accelerated using autonomous coding agents for iterative pair programming. ### How the Agent Accelerated Development:
Security & SSRF Defense Hardening: The agent helped design and stress-test the isSafePublicUrl validator in ssrfDefense.ts, ensuring that the scraper engine cannot be tricked into probing cloud instance metadata endpoints (169.254.169.254) or private RFC1918 subnets (10.0.0.0/8, 192.168.0.0/16).
Deterministic Provenance Verification Engine: To adhere to our "Strict Provenance & Zero Hallucinations" rule, the agent scaffolded the verification pipeline in evidenceEngine.ts. Every single recommendation generated must bind to an audited DOM selector and an observed bounce rate.
Dual-Theme Design System: The agent configured the zero-glow B2B design system, crafting the high-contrast Venom Dark theme (#000000 pitch black with crisp zinc borders) and Spider-Man Light theme with Crimson accents.
Client-Side Vector PDF Architecture: The agent helped calculate pixel-perfect coordinate grids for the 2-page executive audit in exportPdf.ts, ensuring the score gauge, bar charts, and developer checklist render cleanly across any browser without requiring headless Chromium on the backend. > All development commits, schema validations, and agentic workflows are documented directly in the ChurnScope GitHub Commit History. --- ## Prize Categories
Primary Track: Hacktoberfest Weekend Challenge: Build for a Friend

- Partner Categories / Themes: Open-Source AI, Local Inference & Open Innovation, Privacy-First Architecture

Built with ❤️ for Marcus and every indie developer fighting to turn their dream into a sustainable reality.

DEV Community: Talha .

WildSense: The Offline-First AI Nature Companion That Gets You Outdoors

What I Built

Key Capabilities

Demo

Code

2007Talha / Wildsense

🌿 WildSense — An Offline-First AI Nature Companion

✨ Key Features

🧭 1. Adaptive Outdoor Expeditions & Missions

How I Built It

Why Does Open Innovation Matter?

My Agent Session

Prize Categories

What's Next for WildSense

Conclusion & Gratitude

ChurnScope: Saving My Friend’s Startup from Silent User Drop-Off with Open AI & DOM Intelligence

This is a submission for the Hacktoberfest Weekend Challenge: Build for a Friend

What I Built

The Story: Building for Marcus

What ChurnScope Does

Demo

2007Talha / ChurnScope

🔍 ChurnScope

See your website through the eyes of your users.

📖 Overview & The Problem

Architecture & Technology Stack

- The raw data is discarded within 14 milliseconds, retaining only aggregate drop-off percentages.

Why Does Open Innovation Matter?

- Partner Categories / Themes: Open-Source AI, Local Inference & Open Innovation, Privacy-First Architecture