Topic: Ai

The Raw Talks podcast introduces a new round table discussion focusing on practical AI applications. The hosts discuss various ways AI is integrated into their systems, including AI-generated video, audio, and thumbnails for YouTube content. They also touch on using AI for script writing and the concept of "prompting" for clarity.

Pranavi, a co-founder, used ChatGPT to diagnose and find remedies for a headache while fasting. She provided the chatbot with detailed context about her symptoms and fasting status, prompting it to ask further questions to understand her situation accurately. This interaction led to effective remedies within an hour, demonstrating AI's utility in personal problem-solving.

Prompting in AI is defined as clarifying what is desired from the AI model, similar to providing specific details in a conversation to avoid ambiguity. Context engineering, a more advanced concept, involves giving the AI model sufficient background information and tools (like Gmail MCP or Browser User) to effectively solve a problem, rather than letting it figure things out independently.

Perplexity AI is highlighted as a primary tool for deep research, often preferred over ChatGPT, Gemini, or Claude. Its unique "Model Council" feature allows multiple AI models (like GPT, Gemini, and Claude Opus) to collectively research a topic and condense information, providing a more comprehensive and validated output than a single model.

Perplexity's "Model Council" feature, available in its pro version, allows multiple AI models like GPT 5.4, Claude Opus, and Gemini 3.1 Pro to collaboratively discuss and validate a business idea. These models engage in a debate-like process, offering different perspectives and eventually converging on a refined output. A free, manual simulation of this process is also possible by querying individual models and comparing their responses.

Grok Heavy is an AI feature that spins up multiple instances of the same Grok model to approach a problem statement from different angles. Unlike Perplexity's Model Council, which uses distinct AI models, Grok Heavy forces a single AI to think in diverse ways, debate internally, and then converge on a solution. Kimmy has also released a similar version.

AI models were used to validate a business idea for retail optimization, confirming its commercial viability. The models emphasized the importance of using anonymized tracking instead of facial recognition due to India's DPDP Act, and recommended targeting organized restaurant chains first. They also provided a comprehensive analysis and final recommendation, focusing on solving specific problems like customer attention latency and table turnover.

The selection of AI models depends on the specific use case. Perplexity's Model Council is favored for deep research due to its ability to leverage multiple AI biases. For content research, Grok is preferred because of its access to real-time Twitter data, offering diverse perspectives. Gemini 3.1 is used for creative writing and content rewriting due to its superior capabilities in that area.

For creative writing, Gemini 3.1 is highly recommended, while ChatGPT is used for quick, on-the-go tasks like email rewrites. For development, coding, and automation, Claude models (especially Opus 5.4) and OpenAI Codex are preferred, along with Chinese models like Minimax 2.7 accessed via OpenRouter. The importance of selecting the right model for specific tasks, even for non-developers, is emphasized.

OLAMA is a software that allows users to download and run open-source AI models directly on their personal computers or servers, eliminating the need for internet access after initial download. This leverages the computer's GPU for processing. While useful for brainstorming and rewrites, these locally run models should not be used for factual information due to their lack of real-time data access.

Gemma 3, a 12 billion parameter AI model, can be run locally to keep private information safe. While less powerful than larger models, it effectively handles tasks like converting thoughts to notes and generating emails. For enhanced privacy, temporary email generators (burner emails) are suggested. LM Studio and Google AI Edge Gallery offer similar local AI capabilities, with phone-optimized models available for mobile devices.

OpenClaw, an AI employee, is used for extensive personal automation, with access to sensitive data like stock market, bank balances, and social media. Jerry, a version 2.0 built on OpenClaw, integrates with Slack and monitors Twitter, Instagram, and LinkedIn. It provides summaries of meetings, tracks company revenue, analyzes ad spends, and accesses financial dashboards like Razorpay and Stripe, using Julius AI for database interaction.

A common problem for content creators is staying updated with AI and business news. The goal is to automate receiving relevant news updates every three hours via WhatsApp or Slack. This task, while seemingly complex, can be achieved through multiple approaches, with the simplest taking less than half an hour to set up.

AI tools like ChatGPT, Perplexity, and Claude offer a "scheduled task" feature, allowing users to automate recurring actions, such as fetching AI news updates every three hours. This eliminates the need for manual searches. If the AI app is installed on a phone, it can also send push notifications when the task is completed, providing a convenient way to stay informed.

Claude Co-Work is a tool that allows AI to execute tasks and access local system folders, differing from standard chat interfaces that require manual file attachments. This feature is demonstrated by setting up a "Raw AI Update" folder where Claude Co-Work automatically researches AI updates relevant to an infotainment content creator, scores them by relevance, and organizes them into a file, providing real-time progress updates.

Refining AI output requires continuous feedback, guiding the model to improve its relevance and quality over time. The demonstration shows AI generating news updates with content angles for an infotainment creator, including details about Google Vids and Anthropic's Ultra model. This process can be scheduled to run every three hours, updating a new document each time. Advanced integration with APIs like Feedly is also possible for enhanced news sourcing.

For advanced AI news gathering, especially for content creators, relying solely on Google is insufficient. The goal is to access real-time, personalized perspectives from platforms like Twitter and Reddit, which lack direct API access for feed content. This requires giving AI agents the power to open browsers, scroll through personalized feeds, and extract updates, as demonstrated by Jerry, an AI executive assistant built on OpenCloud, which sends curated Twitter updates to Slack.

The Comet browser, an AI-integrated browser, allows an AI agent to monitor personalized Twitter feeds. Since Twitter APIs don't provide access to curated feeds, an OpenClaw-based agent like Jerry can open the Comet browser, scroll through the user's Twitter feed, and extract top tweets related to AI. This automated process, running on a 24/7 computer, sends real-time, non-duplicate updates directly to Slack, providing customized news without manual intervention.

For staying updated on AI, several channels and podcasts are recommended. These include Matthew Berman for technical insights, Andrej Karpathy (ex-OpenAI) for expert perspectives, Lex Fridman's podcast, TVPN (a daily AI news channel recently acquired by OpenAI), and The Moonshot Podcast, which features multiple experts discussing AI three times a week.

GenSpark, a "super app" with AI sheets, can automate the job search process. Users define parameters like job title (e.g., content writing for media agencies/YouTube channels), location (India, remote preferred), expected CTC, required skills, and company type (funded startups, large companies). The AI sheet then intelligently searches and populates relevant job data, streamlining a typically tedious process.