Sunny Patneedi

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

    Current work

    From perspective to practice

    Products, experiments, and field notes from the questions I’m working through. Each case shows my role, a design decision, what you can inspect, and what is still unresolved.

    Browse the archive
    Status: Experiment
    A playbook for parents

    Puddle

    A parent-facing playbook for everyday parenting questions, organised into practical next steps a parent can choose from.

    What does a parent need in the moment: another feed, or a small, concrete next step?

    Audience
    Parents
    My role
    Solo builder of the parent-facing playbook.
    The decision
    Organise guidance as a playbook of practical options a parent can choose from, rather than an open-ended chat that answers for them.
    Inspectable evidence

    The public page at askpuddle.com describes the parent-facing playbook. That page was reached on 9 October 2026.

    A public page is not evidence that the onboarding flow works or that guidance helps a particular family. Puddle makes no medical, developmental or compliance claims.

    Still unresolved
    Which suggestions do parents actually try, and which do they adapt? The intended onboarding flow still needs a dated check before any Available label.
    Verification

    Checked 9 October 2026 · The public page loaded and describes a parent-facing playbook. The intended onboarding flow was not checked, so availability is unverified.

    Source: Public page at https://askpuddle.com. Owner: Sunny Patneedi.

    Earlier capture of the SPEAR repository page
    Earlier capture, not an evaluation result.
    Status: Experiment
    Verification at the runtime boundary

    SPEAR

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

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

    Audience
    Engineers building LLM applications
    My role
    Builder of the open-source LLM security middleware.
    The decision
    Put checks at the runtime boundary and publish an evaluation harness, rather than treating a model provider's safety filters as the whole security boundary.
    Inspectable evidence

    The public repository exposes the source tree, README and evaluation method for readers to inspect.

    Source access is not a released package or hosted service. The documented checks cover input screening and deterministic agent controls, not live-model output leakage or production latency.

    Still unresolved
    What changes with live models, new attacks and real integrations? A bounded evaluation is not a guarantee of complete protection.
    Verification

    Checked 9 October 2026 · The public repository, source tree and README were reachable. This supports View source only; it does not establish a released package, hosted product or supported workflow.

    Source: Public repository at https://github.com/sunnypatneedi/spear. Owner: Sunny Patneedi.

    The project archive

    Current work and earlier experiments. Most archive entries are historical experiments without a dated check for this refresh, so they carry no link and no availability claim. A link appears only where a public destination was verified.

    Available:
    The intended public flow was checked at its exact destination.
    Private preview:
    Access is limited.
    Experiment:
    Prototype or exploratory work; no ready-to-use promise.
    Planned:
    Intent only.
    Status: Experiment
    Current work

    Puddle

    A parent-facing playbook for everyday parenting questions, organised into practical next steps a parent can choose from.

    Checked 9 October 2026

    30 archive entries

    Status: Experiment
    Current work · Web App

    Puddle

    A parent-facing playbook for everyday parenting questions, organised into practical next steps a parent can choose from.

    Audience: Parents

    Checked 9 October 2026

    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.

    Audience: Engineers building LLM applications

    Checked 9 October 2026

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

    SayMake

    An AI-supported maker studio where young people turn an interest into a project they can build, test, explain and improve.

    Audience: Families with young builders

    Not verified for this refresh

    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.

    Audience: Sunny's own workflow

    Not verified for this refresh

    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.

    Audience: People exploring Sunny's earlier experiments

    Not verified for this refresh

    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.

    Audience: People exploring Sunny's earlier experiments

    Not verified for this refresh

    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.

    Audience: People exploring Sunny's earlier experiments

    Not verified for this refresh

    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.

    Audience: People exploring Sunny's earlier experiments

    Not verified for this refresh

    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 visualisation.

    Audience: People exploring Sunny's earlier experiments

    Not verified for this refresh

    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.

    Audience: People exploring Sunny's earlier experiments

    Not verified for this refresh

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

    Stand Up

    A mobile standup timer built with Expo.

    Audience: People exploring Sunny's earlier experiments

    Not verified for this refresh

    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.

    Audience: People exploring Sunny's earlier experiments

    Not verified for this refresh

    Get in touch

    For collaborations, advisory conversations or questions about a project, start on the contact page.