SaaviGenAI LLP

Excited to see this session coming up. Open-weight LLMs give engineers much more control over how models are run, tuned, and evaluated beyond just API usage. Looking forward to the discussion on local models, LoRA, fine-tuning, and practical engineering considerations. If you’re working with GenAI or LLMs, this should be a valuable session.

Learn Open-Weight Models. They are rapidly becoming a trusted choice for enterprise AI adoption, especially for internal AI use cases where privacy, control, and cost matter. Learn from an expert who understands both the technology and real-world enterprise adoption. #openweights #llm #genai #ai

Beyond the API: Engineering with Open-Weight LLMs 🚀 What changes when an LLM stops being just an API—and becomes something you can actually engineer? Most teams experience LLMs the same way: send a prompt, get a response, and optimize around the edges. Open-weight models change that boundary. Join us at SaaviSPARK by SaaviGenAI LLP for an engineering-focused session with Lakshmi Narasimhan, Director of Engineering at Capital One. We’ll explore what becomes possible when you have access to the model itself: ⚙️ Run models locally and understand deployment trade-offs 📦 Quantize and optimize models for practical workloads 🎯 Specialize models using LoRA fine-tuning 🔄 Swap behaviours using adapters 🧩 Compose models and route tasks to larger models when needed 📊 Evaluate whether fine-tuning actually improved performance The session will also include a practical local fine-tuning demo on a Mac, showing how a general-purpose model can be adapted for a narrow task—and how to measure that improvement against the base model using held-out tests. The goal isn’t to argue that open-weight models should replace APIs. It’s to understand when control over the weights creates meaningful engineering value—and when a closed API is still the simpler choice. 📅 9 September 2026 🕖 7:00 PM IST 🌐 Online 🔗 Register here: https://luma.com/sepmhv26⁠� #SaaviSPARK #SaaviGenAI #OpenWeightLLMs #GenerativeAI #AIEngineering #LLM #LoRA #FineTuning #LocalAI #LLMOps #MachineLearning #AIInfrastructure

GenAI data leakage isn't just a policy problem. It's an observability problem. Most orgs try to stop leakage with policy documents and acceptable-use guidelines. But policy can't catch what you can't see. Here's how the exposure actually happens: → Sensitive data enters a prompt → GenAI processes it → The exposure risk goes undetected — because nothing is watching the AI layer → It surfaces later, if at all — as a breach, not a near-miss This is where SaaviAIDR comes in. It monitors AI interactions — prompts, agent activity, and data flows — to give security teams real visibility into where sensitive data moves across GenAI systems, before it becomes an incident. You can't prevent what you can't see. SaaviAIDR brings observability to the AI layer, so exposure doesn't stay hidden until it's too late. Where is sensitive data flowing through your GenAI stack right now — do you actually know? #SaaviGenAI #SaaviAIDR #AISecurity #GenAISecurity #AIObservability #DataSecurity #GenerativeAI #LLMSecurity #AIGovernance #CyberSecurity #AgenticAI #EnterpriseSecurity

AI agents are rapidly becoming digital employees — accessing enterprise tools, data, APIs, and workflows to get real work done. But as agents gain more autonomy, one question becomes critical: Who is observing what they access, what actions they take, and why? This is where SaaviAIDR Observability becomes essential. With better visibility into agent activity, organizations can understand how AI agents interact with systems, monitor permissions and behavior, identify unusual actions, and maintain stronger governance across agentic workflows. Because securing AI agents isn’t only about controlling access. It’s also about having continuous visibility into every important agent action. Identity. Access. Observability. Governance. That’s the foundation for building safer and more accountable AI systems with SaaviAIDR. #SaaviGenAI #SaaviAIDR #AIObservability #AgenticAI #AIAgents #AISecurity #CyberSecurity #AIEngineering #AIGovernance

Eid Mubarak! 🌙 May this Eid bring peace, happiness & prosperity to all. SaaviGenAI The KEY to Secure AI #saavigenai #saaviaide #aisecurity #genai #eid #celebration

Happy Onam! 🌼 May this Onam bring prosperity, happiness & new beginnings. SaaviGenAI The KEY to Secure AI #aisecurity #saavigenai #saaviAIDR #cybersecurity #kerala #keralam #malayalyam

✨ New talent. New ideas. Bigger possibilities. We’re excited to welcome Yasar Ameen to SaaviGenAI LLP as our AI Engineer. Yasar brings a passion for building with Generative AI, AI Agents, LLMs, and emerging AI technologies, and we’re excited to see him contribute to the intelligent solutions we’re creating at Saavigen. 🚀 Here’s to building the future with AI. Welcome to Team Saavigen, Yasar! #Saavigen #AIEngineer #GenerativeAI #AgenticAI #ArtificialIntelligence #NewJoiner #TeamSaavigen

🔐 We’re excited to welcome Atharv Chouhan to SaaviGenAI LLP as an AI Security Engineer! Atharv brings a strong interest in cybersecurity, digital forensics, incident response, network security, and Python, along with hands-on exposure to security investigations and mobile forensics. At SaaviGenAI LLP, he’ll be contributing to our work at the intersection of Artificial Intelligence and Cybersecurity, helping us strengthen secure, responsible, and resilient AI systems. We’re excited to have Atharv join the team and look forward to the ideas, learning, and impact he’ll bring to the journey. Welcome aboard, Atharv! 🚀 #SaaviGenAI #AISecurity #CyberSecurity #ArtificialIntelligence #DigitalForensics #AIEngineer #TeamSaaviGenAI

🇮🇳 Happy Independence Day! Celebrating the spirit of freedom, unity, innovation, and progress. Wishing everyone a proud and joyful 80th Independence Day! Warm wishes from SaaviGenAI LLP. 🇮🇳 Jai Hind! #SaaviGenAI #SaaviAIDR #GenAI #AISecurity #CyberSecurity