A Curriculum Built Around Your Team
Sessions are staged to build fundamentals, extend into hands-on construction, and grow into automated analytical workflows, matched to each team's skill level and application.
Format
Delivered live and in person, so we can observe each participant and confirm concepts stick before moving on. It also gives the team a common framing for tasks and learning, one they carry back to their own workflows together.
Fit to skill level
Sessions are scoped to where a team actually is, not a fixed curriculum. A group new to LLMs starts in Orient; a team already building agents can start in Automate. We mix session composition, by module and by depth, to match individual and team skill level, not the other way around.
Why in person
Hands-on work with life-science prompts asks participants to come in with their own concepts, scopes, and use cases. Sessions are staged so that material is absorbed before the group moves on to the next one.
Why life-science-specific
Every example, exercise, and guardrail is drawn from life-science work: assay data, protocols, LIMS and ELN systems, not generic office tasks. The curriculum spans roles across the lab, digital, analytics, quality, and operations, so teams leave with skills mapped directly onto the tools and data they already use.
Module Topics
Orient, Build, and Automate group the material below. Click any topic to see what it covers.
01 · Orient
Building the fundamentals and understanding of the tools.
LLM Foundations & Mechanics
How large language models generate, reason, and fail, the vocabulary you need before using any tool.
Platform & Model Selection
Comparing providers and platforms against life-science requirements: cost, capability, and data handling.
Data Security & Governance
What data can and cannot go into a prompt, and how your organization's policies apply.
Structured Prompt Engineering
Applying frameworks like ROSE and TAG to get consistent, reliable output.
Development Environment Bridge
Setting up the tools and access that carry training into daily work.
02 · Build
Individuals start constructing tools, skills, and plugins to perform tasks.
Work Product Definition
Naming the deliverable before you build toward it: what "done" looks like.
Spec-Driven Development
Writing a specification the AI can execute against, the same way a protocol governs an assay.
Knowledgebases & RAG Architecture
Grounding outputs in your own documents and data instead of the model's general knowledge.
Verification, Testing & Refinement
Checking AI output against ground truth and iterating until it holds up.
Compliance & Audit Guardrails
Building review and traceability into the workflow from the start.
03 · Automate
Bringing it all together to express your agency with confidence in your current job.
Scheduling & Routine Execution
Running a validated workflow on a recurring basis without a manual restart.
Agentic Loops & Decision Logic
Letting an agent take multiple steps and decide its own next action within set bounds.
Life-Science Platform Integration
Connecting agents to LIMS, ELN, and other systems already in use.
Agent-to-Agent Orchestration
Coordinating multiple agents across a multi-step process.
Human-in-the-Loop Oversight
Defining where a person must review or approve before the agent proceeds.