The network for creativity
Join 1.25M professional creatives like you
Connect with clients, get discovered, and run your business 100% commission-free
Creatives on Contra have earned over $150M and we are just getting started
Back to feedPost
I build the continuity layer that lets AI assistants keep your , rules, and source-of-truth straight across tools and models — without you re-teaching the history every session.
THE PROBLEM
Anyone using more than one AI tool lives the same pain: every new session starts from zero, projects blur together, an old document gets treated as current, and the assistant confidently continues from stale state.
WHAT I BUILT
A project-routing and cross-model continuity architecture (the Global Continuity Router, packaged portably as the VERA Context Bridge):
Explicit project activation and mode persistence
Authority ordering — newest instruction → newest canon → tracker → current conversation
Project isolation, so one project's state doesn't contaminate another
Drive-backed recovery and source-of-truth locations for each project
Hard stops when the authoritative source can't be established, instead of guessing
A portable context package so a different model can reconstruct the working rules and then retrieve current truth, rather than inheriting a stale snapshot
WHAT I CAN DO FOR A CLIENT
Design a continuity / knowledge architecture for a founder, team, or practice juggling multiple AI tools
Organise projects, source-of-truth documents, and trackers so an AI retrieves the right, current version
Build context/onboarding packages that bring a new AI tool or team member up to speed without a week of backstory
I'd version the context package against its Drive sources so a long session can't keep applying rules from an older document. Does the router recheck source revisions before it acts, or only when the session starts?
It rechecks before substantive actions, not only when the session starts. The continuity flow compares the newest dated/versioned canon in Project Sources with the copy in Drive’s LATEST folder, and if they disagree it uses the newer valid version. When the canon itself changes,...
Previously the new source canon was generated in chat and the save state a human directed action, but if the user neglected to save the new canon and/or upload to sources then the model would revert to the previous file. In this instance the model is always comparing the latest...
The network for creativity
Join 1.25M professional creatives like you
Connect with clients, get discovered, and run your business 100% commission-free
Creatives on Contra have earned over $150M and we are just getting started
Related posts
I've built a new mobile AI assistant that brings chat, text and image generation, voice input and document analysis into one app.
It have Smart Chat, Text Creator, Image Create, Voice Input, plus PDF Scanner, Photo Analyze, Social Content and Prompt Ideas, filtered by category.
Almost everything AI can do, just in one place.
Such an amazing tool! Keep going!
Visit the live app here: https://paws-and-puddle-pro.lovable.app/
Meet Sam. He runs a highly requested mobile pet grooming van in Denver, but standard booking calendars are quietly killing his margins. Because generic apps let customers pick any time slot they want, Sam spends two to three hours a day zig-zagging across the city in traffic, burning fuel and losing out on actual appointments just to accommodate a messy schedule.
To fix this, I built Paws & Puddle, a web app that solves his geographic routing nightmare before a booking ever hits the calendar. Instead of a basic form, I engineered a zero-click geo-gate. It automatically fetches a user's zip code via location services and dynamically filters the calendar to only show availability for the specific days Sam’s van is already assigned to their neighborhood.
Because forced logins instantly kill conversion rates, I designed a completely frictionless guest checkout. But to ensure long-term retention, the Thank You page features a "Guest-to-Auth" funnel that seamlessly converts their inputted data into an account post-booking, simply by asking the user to "set a password to track your van's ETA."
Backed by a full operator console where Sam can manage custom service zones, assign route days, and trigger automated status emails, this build proves that premium, minimalist design and aggressive problem-solving belong in the exact same app.
Check out the demo video below to see the end-to-end flow in action! @Lovable #lovablechallenge
Very nice visual design. I love the dog drawings 😍
𝐑𝐀𝐆 𝐀𝐈 𝐊𝐧𝐨𝐰𝐥𝐞𝐝𝐠𝐞 𝐏𝐥𝐚𝐭𝐟𝐨𝐫𝐦 | 𝐈𝐧𝐭𝐞𝐥𝐥𝐢𝐠𝐞𝐧𝐭 𝐒𝐞𝐚𝐫𝐜𝐡, 𝐀𝐈 𝐀𝐧𝐬𝐰𝐞𝐫𝐬 & 𝐕𝐞𝐜𝐭𝐨𝐫 𝐃𝐚𝐭𝐚𝐛𝐚𝐬𝐞
I designed and built a RAG-powered AI knowledge platform that lets businesses search documents, websites, databases, and internal knowledge using natural language.
The system processes content, creates embeddings, stores them in a vector database, retrieves the most relevant information, and uses AI to generate accurate, source-grounded answers.
My services include: RAG development, document ingestion, semantic search, vector database setup, OpenAI/LLM integration, internal knowledge assistants, API integrations, and analytics.
The solution helps teams find information faster, reduce repetitive research, improve answer consistency, and build scalable AI-powered knowledge systems.
𝐓𝐞𝐜𝐡 𝐒𝐭𝐚𝐜𝐤:
OpenAI, LangChain, RAG, Pinecone, Qdrant, Weaviate, FAISS, Chroma, Python, FastAPI, PostgreSQL, APIs.
