

EP 590: Agents, LLMs, or Algorithms? A Playbook for Choosing AI
79 snips Aug 15, 2025
Michael Abramov, CEO of Keymaker and Key Labs, dives deep into the world of AI. He demystifies the differences between algorithms, large language models, and agentic AI, offering actionable insights for leaders. Abramov emphasizes the importance of data in AI training and discusses the risks and benefits of adopting agentic AI. He also highlights the necessity of experimentation in AI for business innovation, urging organizations to balance routine practices with creative problem-solving. Tune in for a compelling exploration of AI's evolving landscape!
AI Snips
Chapters
Transcript
Episode notes
Block Time To Experiment With AI
- Block time daily to experiment with AI tools and learn by doing.
- Use personal pain points as the first places to test new AI workflows.
Treat LLMs As Black-Box Tools
- LLMs act like a flexible black box that answers many types of prompts.
- Treat models as tools, not entire systems, when designing solutions.
Agents Orchestrate Multiple Models
- Agents are orchestrations of multiple models connected with conditional flows.
- They coordinate checks and tasks by routing between models using if/else logic.