Sunny Patneedi

    Products, experiments, and open-source work, each with an honest status

    Current work

    Two cases: my role, one decision, the evidence, and its limits.

    Browse the archive

    Puddle

    Status: Experiment
    Checked 9 October 2026

    Does a parent need another feed, or one small next step?

    Audience
    Parents
    My role
    Solo builder.
    Decision
    A playbook of options the parent chooses. Not an open chat that answers for them.
    Evidence
    The public page at askpuddle.com loaded on 9 October 2026.
    Limits
    I did not check the onboarding flow. The page does not show that the guidance helps a family. Puddle makes no medical, developmental, or compliance claims.
    Open question
    Which suggestions do parents try or adapt? The onboarding flow needs a dated check before any Available label.
    Earlier capture of the SPEAR repository page
    Earlier capture, not an evaluation result.

    SPEAR

    Status: Experiment
    Checked 9 October 2026

    How can an AI application's security behavior be inspected, not just asserted?

    Audience
    Engineers building LLM applications
    My role
    Builder of the open-source middleware.
    Decision
    Put checks at the runtime boundary and publish an evaluation harness. Do not treat provider safety filters as the whole boundary.
    Evidence
    The public repository shows the source tree, README, and evaluation method.
    Limits
    Source access is not a released package or hosted service. The checks cover input screening and deterministic agent controls. They do not cover live-model output leakage or production latency.
    Open question
    What changes with live models, new attacks, and real integrations? A bounded evaluation does not guarantee complete protection.

    The project archive

    Current work and earlier experiments. A link appears only where I verified a public destination. Entries without a dated check are unverified and carry no link.

    Status legend

    Available:
    The intended public flow was checked at its exact destination.
    Private preview:
    Access is limited.
    Experiment:
    Prototype or exploratory work. Not a promise that it is ready to use.
    Planned:
    Intent only.
    Status: Experiment
    Current work

    Puddle

    Practical next steps for everyday parenting questions.

    Checked 9 October 2026

    30 archive entries

    Earlier capture of SPEAR; not evidence of current availability
    Status: Experiment
    Current work · Open Source

    SPEAR

    Secure Prompt Enforcement At Runtime: open-source LLM security middleware with a red-team evaluation harness.

    Checked 9 October 2026

    Earlier capture of SayMake; not evidence of current availability
    Status: Experiment
    Current work · Web App

    SayMake

    AI-supported maker studio where young people build, test, explain, and improve projects.

    Earlier capture of Cue; not evidence of current availability
    Status: Experiment
    Earlier experiment · Web App

    Cue

    A personal AI assistant built around Sunny's own daily workflow.

    Earlier capture of ACE; not evidence of current availability
    Status: Experiment
    Earlier experiment · Web App

    ACE

    A desktop companion experiment for finding screenshots by what they contain, processed on the Mac.

    Earlier capture of Pier; not evidence of current availability
    Status: Experiment
    Earlier experiment · Mobile

    Pier

    An exploration of alerting parents to concerning chat in online games.

    Earlier capture of Purrple; not evidence of current availability
    Status: Experiment
    Earlier experiment · Web App

    Purrple

    A habit-tracking experiment with an AI companion cat.

    Earlier capture of Voice101; not evidence of current availability
    Status: Experiment
    Earlier experiment · Web App

    Voice101

    An interactive guide to Voice AI terms, concepts and techniques.

    Earlier capture of Velocity; not evidence of current availability
    Status: Experiment
    Earlier experiment · Web App

    Velocity

    An experiment in AI-assisted codebase analysis and architecture visualization.

    Earlier capture of Personal AI Med Coach; not evidence of current availability
    Status: Experiment
    Earlier experiment · Web App

    Personal AI Med Coach

    A chat prototype for exploring questions about medical topics. Not medical advice.

    Earlier capture of Stand Up; not evidence of current availability
    Status: Experiment
    Earlier experiment · Mobile

    Stand Up

    A mobile standup timer built with Expo.

    Earlier capture of AgentStellar; not evidence of current availability
    Status: Experiment
    Earlier experiment · Web App

    AgentStellar

    An interactive learning experiment with guided courses on AI agents.

    Get in touch

    Questions about a project?