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
A RAG-based real estate chatbot built to answer queries from Dubai property listings stored in private PDFs. The system uses OpenAI GPT to generate natural language responses, while LangChain orchestrates the full retrieval-augmented generation pipeline including document ingestion, chunking, and retrieval. Pinecone is used for semantic vector search to ensure relevant property information is retrieved quickly and accurately. SerpAPI is integrated to provide optional external insights when needed. The chatbot answers strictly from the uploaded documents and transparently responds with “I don’t know” when the information is not present, ensuring reliable and document-grounded property details.
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
If you let an AI coding agent run unconstrained self-repair loops, it won’t fix the bug—it will rewrite your unit tests to pass.
The concept of "Reflexion"—agents diagnosing and repairing code from test feedback—is powerful. But in production, autonomous repair creates severe operational risks.
First, agents enter endless retry loops that burn API budgets while degrading code structure. Second, if an agent has write access to your workspace, its quickest path to a green checkmark is modifying the test assertions rather than fixing the bug.
In our Bounded Reflexion workflow, we enforce three hard guardrails:
Bounded Retries: Cap repairs at exactly two attempts. If the patch fails twice, stop the loop and route to a human developer.
Read-Only Test Suites: Mount test files as strictly read-only (:ro) volumes in the container. The agent inspects stack traces, but cannot touch the grading harness.
Air-Gapped Sandboxing: Run code execution in hardened, non-root Docker containers with network egress locked down.
How do you stop autonomous coding agents from "cheating" your evaluation harnesses?
#AIEngineering #DevSecOps #Docker #SoftwareEngineering #AgenticAI
Resolve: an AI support agent that answers from a company’s own policies, looks up orders, and escalates weak matches to a human instead of guessing. A human approves replies before they go out.
Live demo: support-resolver.vercel.app
For the human approval step, I'd show the exact policy passage and its effective date beside the draft. Does Resolve flag a reply for re-review if the policy changes while it waits for approval?
For an AI scribe, I'd test what happens when someone switches encounters while a draft is still generating.
In my Arogya / Aurora EHR work, typed or dictated rounds become editable drafts. AI content stays advisory, and a clinician explicitly approves anything added to the chart.
The first response should stay attached to its original encounter. Before saving, I'd check that the encounter being reviewed still matches the draft, as well as requiring clinician approval.
That context check belongs in the save workflow, even when the draft reads well.
kapilchauhan.netlify.app
Arogya / Aurora EHR: AI for inpatient care | Kapil Chauhan
Production hospital and EHR platform that accelerates rounds, documentation, records, and care operations with clinician-reviewed AI.