AI Lab

Sandbox for prompt engineering libraries, model benchmarks, and active agent workflows.

Model Comparison Matrix

Benchmarks of top LLMs for developer automation, code formatting, and speed-to-cost scales.

Benchmark Feature	GPT-4o (OpenAI)	Claude 3.5 Sonnet	Gemini 1.5 Flash	Llama 3 (Meta)
Primary Use Case	Complex reasoning, writing, coding	Deep analysis, coding, long-context tasks	Fast speed, multimodal input, news curation	Local self-hosted applications, privacy
Context Window	128K tokens	200K tokens	1 Million+ tokens	8K - 128K tokens
Speed / Cost	Medium / Premium	Medium / Premium	Extremely Fast / Ultra Low Cost	Variable (Depends on self-host hardware)
Code Quality	Excellent (Very direct)	Outstanding (Best structure & comments)	Very Good (Reliable syntax)	Good (Requires careful prompts)

AI Ingestion Pipelines

How autonomous agents extract, format, and publish news updates on StartupWire.in.

STAGE 01

Ingestion Worker

Cron trigger initiates every 4 hours. Node scripts fetch RSS XML payloads from pre-audited tech sites and map schemas in memory.

STAGE 02

LLM Validation

Gemini 1.5 Flash cleans duplicate nodes via vector cosine comparisons. Feeds are summarized based on strict prompt constraints.

STAGE 03

Database Commit

Formats output as JSON Schema. Stores payload inside Supabase PostgreSQL, prompting automated sitemap regeneration.

Prompt Engineering Library

Optimized prompts designed for development workflows, copywriting, and security decodes.

SEO Article Writing Assistant

SEO

Generates SEO-friendly tech blogs with headings, metadata, and appropriate JSON-LD schema layouts.

SYSTEM INSTRUCTION:

You are an expert tech writer and SEO specialist. Write content that is accurate, factual, readable, and highly optimized for crawlers.

USER PROMPT PLAYLOAD:

Act as a senior technology writer. Write a comprehensive guide on the topic: {{topic}}.
Requirements:
1. Include an H1 title and H2/H3 subheadings.
2. Provide a meta description (under 160 characters).
3. Draft the article in clean markdown.
4. Keep the tone professional, educational, and engaging.
5. List 3 key keywords to target.

Contains variable bindings (e.g. {{var}})

Clean Code & Type-Safety Refactorer

Refactoring

Refactors JavaScript/TypeScript code to maximize performance, clean structure, and robust type safety.

SYSTEM INSTRUCTION:

You are a principal software engineer. You value type safety, clean code principles, readability, and performance. Do not output explanations unless asked.

USER PROMPT PLAYLOAD:

Analyze the following code snippet and refactor it:
```typescript
{{code_snippet}}
```
Refactoring Rules:
- Ensure all types are explicitly defined.
- Optimize loops and asynchronous calls.
- Implement proper error handling.
- Keep helper functions modular.

Contains variable bindings (e.g. {{var}})

CTF Log Decoder

Debugging

Decodes hex/base64 representations and performs preliminary security vulnerability checks.

SYSTEM INSTRUCTION:

You are a cybersecurity analyst. Help analyze CTF challenge logs without giving direct flags, guiding the learning process.

USER PROMPT PLAYLOAD:

Analyze this log snippet:
{{log_text}}
Identify the potential vulnerability category, suggest 3 investigation commands, and describe how to avoid this threat in code.

Contains variable bindings (e.g. {{var}})