Company

Team

Before building LLM Council, Ash Tiwari worked on two research programs at Arizona State University’s W.P. Carey School of Business: a human consumer-choice study comparing peer and anonymous restaurant reviews with Dr. Timothy J. Richards, and an econometric study of spatial patterns in green-building certification with Dr. Yueming (Lucy) Qiu. That work inspired an interest in network effects; it did not study LLMs or prove that AI councils outperform single models.

From research to product

The restaurant study found peer networks substantially more effective at shaping restaurant preferences than anonymous reviews after controlling for endogeneity; it did not measure whether group decisions were more accurate. The building study found strong spatial correlation in certification diffusion.

LLM Council applies a product-design analogy: independent perspectives can be compared and dissent can be surfaced. Direct support for multi-model methods comes instead from recent LLM experiments, whose results are specific to the tasks, models, prompts, and aggregation methods tested.

Ash Tiwari — Founder

Certified AI engineer and NLP researcher. Builds and scales production-grade AI systems for government and enterprise. Research background in predictive modelling and discrete choice models (Arizona State), deep learning and NLP (Indian School of Business), and LLM fine-tuning (QUT). 73+ citations on Google Scholar.

LinkedIn: https://www.linkedin.com/in/ashtiwarievolo/ · GitHub: https://github.com/ashtiwariasu · Scholar: https://scholar.google.com/citations?user=5WLe96EAAAAJ · ash@llmcouncil.ai

Experience

Sixteen years across regulated government and enterprise environments:

  • 2025 — Department of Customer Service, Queensland: Qchat (35,000 government users, 7,000 daily active) and Corella, the student and teacher AI assistant rolling out statewide in 2026
  • 2024–2025 — Energy Queensland: Microsoft 365 Copilot rollout for 3,000 users; Azure AI Foundry and AWS Bedrock platforms
  • 2023–2024 — Department of Education, Queensland: RAG-based prioritisation engine on Databricks; analytics architecture on Azure Synapse and Fabric
  • 2022–2023 — Department of Foreign Affairs and Trade: multi-country platform for Pacific Labour Mobility
  • 2020–2023 — QUT: fine-tuning GPT-3 and BART for context-aware summarisation of social media conversations
  • 2019–2020 — Indian School of Business: AI applications and policy research
  • 2016–2019 — Founder and CTO, AN Services (now Scanxt, India)
  • 2013–2016 — American Express, Phoenix AZ: enterprise architecture and analytics for card services
  • 2011–2013 — Arizona State University: predictive analytics and discrete choice demand models

Publications and talks

Research areas: transformer architecture, NLU/NLG fine-tuning, multi-agent consensus systems, social network analysis.

December 2025 — CSIRO Responsible AI Symposium, Adelaide: “Grounding Global Standards in Australian AI Governance”.

Full publication list: https://scholar.google.com/citations?user=5WLe96EAAAAJ

Certifications

  • Azure AI-102 (AI Engineer), AI-900, DP-900
  • ISO 42001 Lead Implementer
  • TensorFlow, NLP and CNNs — Google Brain / DeepLearning.ai
  • Certified Scrum Master

LLM Council home · Pricing · How it works