§ 04 · Practice
Agents that do the work, and show their work.
We build agents that operate inside your existing tools: they launch pipelines, read QC reports, curate literature, draft documents and hand off to a person at the points you choose. Every action is logged and reversible.
§ 04.1 / Scope
Capabilities.
Analysis operators
Agents that launch Nextflow runs, watch for failures, retry sensibly and summarise results when a run completes.
QC triage
Read MultiQC and pipeline metrics, flag outlier samples, propose re-runs, and explain why in plain language.
Curation assistants
Literature screening, variant curation drafts, and structured extraction from papers into your schema.
Report drafting
Structured outputs into your templates: methods, results, figure captions, with links back to the data.
Tool integration
Model Context Protocol servers for your LIMS, ticketing, Slack, cluster and cloud, with scoped permissions.
Safety & oversight
Approval gates, budgets, rate limits, full trace logs, and evaluation and red-teaming before anything touches production.
§ 04.2 / Output
Deliverables.
- —Agents with defined tools, permissions and escalation paths
- —Trace logs and dashboards for every action taken
- —An evaluation suite and a staged rollout plan
- —Training for the team that will run them
Tools & Standards.
nf-core and community standards where they exist, custom software where they do not.
- Claude Agent SDK
- MCP
- LangGraph
- Pydantic
- Temporal
- OpenTelemetry
- Nextflow
- Slack
§ 04.3 / Questions
Frequently asked questions.
- What can an agent be trusted to do on its own?
- We start with read-only and draft-only tasks, then expand autonomy as the evaluation numbers and your team's confidence allow. Destructive actions always require approval.
- Do agents need cloud access to our data?
- No. Agents can run entirely on your infrastructure, including with on-device models, so no data has to leave.
§ Practice Scoping · 04
Scope a project in ai agents.
Tell us about your assay types, dataset scale, and timelines. We reply with a preliminary scoping memo within two working days.
Scope AI Agents →