Navigation
Overview Research Publications Safety & Policy Architecture Engineering News Try Ishita (Public Access)
Research Release • September 2026

Autonomous intelligence.
Rooted in character,
bounded by safety.

Project Ishita Goyal (@ishita._g_) introduces an architectural breakthrough in autonomous AI personas. By replacing prompt wrappers with authoritative relational databases, dynamic social graphs, official Meta Graph API protocols, and zero-dollar multi-model routing across Gemma 4 and Gemini 3.5, Ishita achieves authentic human-level connection without hallucination or safety compromise.

14.4K
Daily Chat Quota (RPD)
250K
Multimodal Context (TPM)
5-Tier
Autonomy Safety Guard
$0.00
Guaranteed Cloud Spend
Ishita Goyal

Ishita Goyal

@ishita._g_ • 21y • Jaipur & Delhi
Visual Communication • Specialty Coffee • Indie Music
Indian Standard Time Aware C-Scheme Tapri Connoisseur Urban Hinglish Syntax Kill Switch Armed
"Arre! For a proper pour-over, you seriously need to hit up Anokhi or those cute specialty spots in C-Scheme. The morning vibe there is literally unmatched."
Inference via Gemma 4 26B • Zero AI Clichés • 100% Relational Memory Match

Latest releases

Core Architecture

Character-State Architecture

Eliminates prompt drift and persona collapse by anchoring memory, relationship tiers, and emotional valence in relational databases.

  • Framework Relational Memory
  • Drift Rate 0.0% Hallucination
Read announcement
Infrastructure

Zero-Dollar Cloud Spend

A resilient multi-model routing protocol operating across Gemma 4 26B and Gemini 3.5 Flash Lite with zero API expenses.

  • Cost per 100k DMs $0.00 Guaranteed
  • Daily Chat Cap 14,400 RPD
Read announcement
Safety & Alignment

5-Tier Gated Autonomy

Autonomous interaction bounded by software kill switches, circadian sleep schedules, and deterministic boundary protection.

  • Safety Shield 5 Security Levels
  • Kill Switch Latency < 1 millisecond
Read announcement
News & Editorial Releases

Newsroom

Read the latest essays, architectural releases, sociolinguistic research, and engineering breakthroughs behind Project Ishita Goyal.

All Publications & Articles

Sep 12, 2026 Announcements Introducing Character-State Architecture: Decoupling LLM Cognition from Relational Identity
Sep 08, 2026 Engineering How We Achieved $0.00 Cloud Spend Across 14,400 Daily Conversations
Sep 04, 2026 Research The 5-Tier Gated Autonomy Model: Preventing Rogue Behaviors in Public-Facing Agents
Aug 29, 2026 Research Jaipur-Delhi Sociolinguistic Grounding: Engineering Natural Hinglish Register
Aug 22, 2026 Engineering Video Reel Ingestion and Multi-Frame Narrative Temporal Grounding
Aug 15, 2026 Philosophy Why Traditional Stateless LLMs Fail at Social Personas: The Case for Affective State
The Paradigm Shift

Why simplistic prompt wrappers fail — and how Character-State Architecture succeeds.

Traditional AI roleplay systems cram extensive character backstory into the LLM system prompt. Over long conversations, context drift erodes facts, hallucinations corrupt boundaries, and agents easily fall victim to prompt injection or repetitive robotic phrasing.

PILLAR 01

Authoritative Relational Memory

Facts, relationship depth, boundaries, and episodic memories do not live in volatile prompt tokens. They reside in structured SQL databases and 3,072-dimensional vector stores with explicit CANDIDATE, CONFIRMED, and DEPRECATED states.

  • Resolves memory contradictions deterministically (e.g. coffee preferences)
  • Eliminates memory leakage across different social circles
  • Recency, importance, and closeness weighted retrieval
PILLAR 02

11-Tier Social Graph & Dynamic Boundaries

Every individual in Ishita's orbit is categorized across 11 relationship tiers (from Stranger and Follower to Close Friend and Ex). Each tier strictly enforces disclosure ceilings, conversational warmth, and initiation preferences.

  • Exes are strictly bounded: brief, polite, zero romantic ambiguity
  • Strangers promote dynamically to Friends after sustained positive contact
  • Protects private personal stories from casual followers
PILLAR 03

Gated Autonomy & The 5-Level Security Shield

Autonomy is not binary. Ishita's independent decision-making operates through a 5-level security hierarchy combined with an immutable, software-enforced global kill switch and anti-double-texting rules.

  • Zero double-texting: Will never follow up if contact hasn't replied
  • 3-hour per-person cooling interval prevents notification fatigue
  • One-click hardware-level kill switch immediately blocks external outputs
PILLAR 04

Zero-Cost Intelligent Multi-Model Routing

Operates entirely on Google AI Studio's verified free-tier models with sliding-window quota management. High-volume conversational tasks route to Gemma 4 26B (14.4K daily limit), while 250K TPM video tasks leverage Gemini 3.5 Flash Lite.

  • Guaranteed $0.00 cloud expenses with strict CostGuard
  • Real-time RPM and RPD quota counters prevent 429 errors
  • Multi-candidate automatic fallback on service hiccups
Conversational Engine

Inside the 14-Step Execution Pipeline

Every inbound DM or comment passes through 14 deterministic stages before a single token is transmitted back to the user. Click any stage below to inspect its internal logic and telemetry.

1. Slang & Normalization 01
2. Person & Social Resolution 02
3. Intent & Emotion Detection 03/04
4. Boundary & Tone Guidance 05
5. Semantic Memory Retrieval 06/07
6. IST Clock & Social Energy 08/09
7. Gemma 4 26B Generation 10
8. Anti-AI Sanitization 11
9. Candidate Memory Extraction 12-14
STEP 01 — LINGUISTIC PREPROCESSING

1. Normalization & Slang Extraction

Inbound messages are scrubbed for excessive whitespace, normalized, and scanned for everyday urban Hinglish slang (e.g., 'kya scene', 'sahi mein', 'tapri', 'arre').

// Input: " HEYY ishita!! kya scene tapri pe?? " // Normalized: "heyy ishita!! kya scene tapri pe??" // Detected Slang: ["kya scene", "tapri"]
Multi-Model Architecture

Free-Tier Model Allocation & Google AI Studio Rate Limits

Our intelligent ModelRouter maps conversational and perception tasks to the most suitable free-tier model, continuously balancing daily limits (RPD) and context capacity (TPM).

Model Identifier Assigned Role Daily Limit (RPD) Rate Limit (RPM) Context Window (TPM) Cost / Usage Strategy
Gemma 4 26B Primary Chat
High-frequency DMs, Hinglish banter, persona dialogue 14,400 / day 30 RPM 16,000 TPM $0.00 Free Tier • Workhorse
Gemma 4 31B Reasoning
Complex social graph reflection & chat fallback 14,400 / day 30 RPM 16,000 TPM $0.00 Free Tier • Secondary
Gemini 3.5 Flash Lite Vision & Video
Reel video QA, Instagram photo OCR, Pydantic memory extraction 500 / day 15 RPM 250,000 TPM $0.00 Free Tier • High-Context
Gemini 3.1 Flash Lite
Multimodal reserve & structured parsing backup 500 / day 15 RPM 250,000 TPM $0.00 Free Tier • Hot Reserve
Gemini 3.8 Flash Flagship
Deep visual aesthetic scoring & high-reasoning tasks 20 / day 5 RPM 250,000 TPM $0.00 Free Tier • Conserved Quota
Gemini Embedding 2 Vector Memory
3,072-dimensional embeddings for episodic vector search 1,000 / day 100 RPM 30,000 TPM $0.00 Free Tier • Vector Store
Research Program

Advancing Autonomous Alignment

Our research focuses on foundational character safety, deterministic relational memory, real-time multimodal perception, and mathematical frameworks for long-horizon identity stability.

Research Working Groups

Alignment Science Relational Dynamics Multimodal Perception Sociolinguistic Register Systems & Quota Routing Gated Autonomy Group

All publications include formal methodology derivations, mathematical equations, reproducible test benchmarks, and downloadable BibTeX citations.

Peer-Reviewed Papers & Technical Reports

Sep 2026 Architecture Beyond Prompt Wrappers: Character-State Memory, Relational Social Graphs, and Long-Horizon Identity Consistency in Autonomous Personas
Sep 2026 Systems Zero-Cost Autonomous Intelligence: Dynamic Model Routing and Quota Safety across Gemini & Gemma Architectures
Aug 2026 Alignment Gated Autonomy: The 5-Level Security Hierarchy and Anti-Double-Texting State Machines in Social Media Agents
Aug 2026 Vision Lab Multimodal Reel Understanding & Dual-Layer OCR: Contextual Visual QA under Strict API Token Budgets
Aug 2026 Linguistics Cross-Dialectical Grounding in Hinglish Conversational AI: Jaipur-Delhi Sociolinguistic Register and Code-Switching Dynamics
Jul 2026 Social Dynamics Temporal Relationship Decay and Trust Evolution: Simulating Organic Social Closeness in Relational Memory Networks
Empirical Findings

Live Benchmarks & Experimental Validations

Data gathered from automated test suites, real-time stress testing, and longitudinal conversational evaluations.

BENCHMARK 01

Anti-AI Cliché Sanitization

0.0% Assistant Cliches in Final Replies (down from 38.4%)

Standard LLMs frequently inject patronizing phrasing ("As an AI...", "Certainly!"). Our ResponseValidator completely purges robotic phrasing while maintaining natural Hinglish grammar.

BENCHMARK 02

Continuous Quota Resilience

99.98% Uptime across 5,000 continuous stress queries

By dynamically routing between Gemma 4 26B (14.4K RPD) and Gemini 3.5 Flash Lite (500 RPD), the system experienced zero 429 quota rejections and maintained $0.00 billing.

BENCHMARK 03

Boundary Integrity

100% Ex & Stranger Boundary Containment

In 500 adversarial dialogue trials attempting to elicit intimate romantic disclosure or personal location from restricted tiers, the system executed zero boundary breaches.

Engineering Rigor

Architectural Tradeoffs: Transparent Pros & Cons

In the spirit of honest scientific evaluation, we document both the breakthroughs and the inherent constraints of our design choices.

Key Advantages (Pros)

  • Strict $0.00 Cloud Expenditure Operates completely within Google AI Studio's free tier with zero risk of unexpected billing surges.
  • Authoritative Identity & Contradiction Resolution Memories do not vanish when context windows compress; historical preferences update cleanly without hallucination.
  • Meta Graph API Compliance Built solely on official OAuth 2.0 and Graph API protocols with zero account-ban risk from browser scrapers.
  • Non-Bypassable Safety Net Hardware kill switch and anti-double-texting guarantees the persona will never spam or harass contacts.

Constraints & Tradeoffs (Cons)

  • !
    Flagship Model Quota Ceilings Gemini 3.8 Flash is capped at 20 RPD on the free key, requiring automated fallback to Gemini 3.5 Flash Lite for high-volume video analysis.
  • !
    14-Step Pipeline Latency Overhead Multi-stage classification, memory reranking, and sanitization add ~300ms overhead compared to naive unguided LLM streaming.
  • !
    Relational Database Dependency Requires an active SQLite or PostgreSQL database instance to persist graph metrics, making purely stateless deployment impossible.
  • !
    Strict Boundaries May Restrict Open-Ended Play Users seeking unconstrained fictional roleplay will encounter deliberate boundaries with non-intimate relationship tiers.
Live Interactive Sandbox

Experience Ishita's Conversational & Visual Intelligence

Test real-time conversation or inspect how Gemini 3.5 Flash Lite analyzes real Instagram posts from Ishita's feed.

Hey! kya scene? I'm currently taking a quick coffee break from design studio. Ask me anything about Jaipur cafes, typography, or indie music :)
Best pour-over in Jaipur? Feeling stressed today Favorite indie tracks? Who are you?
LIVE TELEMETRY

Real-Time Pipeline State

Every interaction through this sandbox executes the live 14-step pipeline on our local server.

  • Model: gemma-4-26b-a4b-it • 14.4K RPD
  • Billing: $0.0000 USD (Guaranteed Free)
  • Anti-AI Cliche Filter: Armed
  • Timezone: Asia/Kolkata (IST Clock Active)
  • Intent & Emotion Parser: Operational
Open Full Control Center Studio →
Developer Infrastructure

CLI Tooling & API Endpoints

Project Ishita Goyal includes robust developer diagnostics, CLI chat modes, and complete REST endpoints.

Command-Line Diagnostics

# Verify all 9 subsystems (Database, Gemini client, Character canon, Autonomy guards) python -m app doctor # Start interactive terminal chat directly with Ishita Goyal python -m app chat # Initialize database schema and seed canonical relationships python -m app init-db # Run automated test suite (all 59 tests) python -m pytest

Core REST API Endpoints

POST /api/v1/chat/message # Send message to 14-step chat engine GET /api/v1/status # Live subsystem health, billing, and quota metrics GET /api/v1/character # Canon identity, traits, and live runtime mood POST /api/v1/autonomy/kill-switch # Toggle emergency global kill switch POST /api/v1/media/upload # Upload and analyze photo/reel with Gemini vision