What you will be able to show
A small tool-using agent with an evaluation set, permissions boundary, and failure-handling plan.
ai · Intermediate
Agentic Software Engineering
Design bounded agents that use tools, manage context, recover from failure, and pass measurable evaluations.
Learn activelyShort decisions before explanation
Verify independentlyTests, sources, and failure cases
Finish with proofArtifacts another person can inspect
Skills practiced in context
Four-mission proof sprint
Do the work. Keep the evidence.
Check a mission only after the proof exists. Progress stays in this browser and also appears in the Modern Skills catalog.
-
01
Choose the right boundary
Separate deterministic code, model judgment, tool calls, and human approval.
Proof: A workflow diagram with stop and escalation conditions. -
02
Manage context deliberately
Create compact instructions, task-specific memory, and a policy for what never enters context.
Proof: A context budget and data-handling checklist. -
03
Connect one safe tool
Define a narrow tool contract, validate inputs, and make retries idempotent.
Proof: A tested tool schema with denied unsafe inputs. -
04
Evaluate the loop
Build ten representative cases including failure, ambiguity, and adversarial input.
Proof: An eval report with pass rate and one improved failure.
0 of 4 proof missions complete
0%
Proof sprint complete
Explore $29 founding access
Want structured feedback and deeper practice?
The free path remains yours. Founding access helps shape future mock interviews, review rubrics, and portfolio verification.
Primary learning sources
Go deeper without entering an endless feed.
These external resources are optional benchmarks. Complete the LearnByShorts mission and bring back one useful artifact.