LOCKEDIN / FORWARD-DEPLOYED ENGINEERING 101
Understand it.
Then build it.
You can be an experienced engineer and still be new to agents. Start with what each thing is and how it helps a person do useful work. Explore 64 original introductions across software, agents, cloud, security and delivery. Start here without a model key; each topic connects to deeper instruction and practice.
A place to start.
Choose a short reading route, or look up a concept below.
- 101UnderstandPlain language, visual walkthrough, vocabulary
- 102Build with guidanceOriginal lessons, examples and knowledge checks
- 103Practise and explainRunnable labs, failure cases and handover
Recovery engineering · data and service objectives
What do RTO, RPO and failover mean?
RTO is the target maximum service interruption. RPO is the target maximum age of recoverable data relative to the disruption. Restore recovers from a saved point; failover moves service to a replacement. These are related but different responsibilities.
An FDE must explain what work can be lost, how long the business can wait and what proves the receiving service is usable.
UNDERSTAND THE SYSTEM
Recovery has two different clocks
RPO limits the age of the data you can afford to lose. RTO limits how long the business can wait for an agreed usable service. Both are objectives to test.
Read the connections in plain English
- 01Recoverable point → Later accepted work
A backup cannot contain later transactions unless another recovery mechanism captures them.
- 02Later accepted work → Service interruption
Count actual missing records and the age of the recoverable point separately.
- 03Service interruption → Restore and validate
Include detection, decision and preparation in the declared recovery timeline.
- 04Restore and validate → Service accepted
An open port or a restored database alone may not satisfy acceptance.
- 05Service accepted → Compare with objectives
Compare the 12-minute interruption with the 15-minute target.
- 06Recoverable point → Compare with objectives
Compare the recovery-point age at interruption with the 5-minute target.
Check your understanding: Is a replica a complete backup strategy?
No. Replication can also copy accidental deletion or corruption. Failover, retained recovery points and a tested restore solve different failure cases. Promotion must prevent conflicting writers and preserve a deliberate failback plan.
CONNECT IT TO WHAT YOU KNOW
Start with something familiar.
Recovering a saved document and reopening the application are separate jobs. One governs which work survives; the other governs when the service can be used.
A fast restore can still lose too much work. A replica may copy accidental deletion. Measure the useful path, not only process restart.
Words worth knowing
- RTO
- Recovery Time Objective: target service interruption duration.
- RPO
- Recovery Point Objective: target maximum recoverable-data age at disruption.
- Failover / failback
- Move service to a replacement / deliberately return later.
- Fencing
- Prevent a former owner from continuing authoritative writes.
- Backup / replica
- A retained recovery point / another copy following changes; they solve different failures.
Useful foundations
Open a definition and diagram here. Your lesson stays in place.
Data foundations · object storage, SQL and retrievalFiles, databases and search indexes: what goes where?
QUICK REFERENCE / 101
Files, databases and search indexes: what goes where?
An application needs somewhere to keep information and a way to find it again. Object storage keeps named objects such as documents and images. A relational database organizes records into tables you can query and update. A search index helps find relevant material. These are different jobs, even when one product can do more than one of them.
An FDE needs to explain where the customer's information lives before connecting it to an agent. A convincing answer is not enough if it used an outdated procedure, lost a ticket update or crossed a customer's access boundary.
A deployment and operating environmentWhat changes when AI runs in a cloud?
QUICK REFERENCE / 101
What changes when AI runs in a cloud?
A cloud platform provides managed computing, storage, networking and identity services. A managed model endpoint is one service inside that environment. Building AI in a cloud means connecting these pieces into an application with explicit data, access, operational and cost boundaries.
An FDE must design for the customer's actual environment rather than assume every provider exposes the same behavior.
ONE EXAMPLE / FICTIONAL CASE
Follow the connections.
A fictional backup covers 10:00. Interruption starts at 10:08 and service is accepted at 10:20. The data gap is eight minutes and interruption twelve. With RPO five and RTO fifteen, only the time target is met.
UNDERSTAND THE SYSTEM
Recovery has two different clocks
RPO limits the age of the data you can afford to lose. RTO limits how long the business can wait for an agreed usable service. Both are objectives to test.
See what moves and why.
Read the connections in plain English
- 01Recoverable point → Later accepted work
A backup cannot contain later transactions unless another recovery mechanism captures them.
- 02Later accepted work → Service interruption
Count actual missing records and the age of the recoverable point separately.
- 03Service interruption → Restore and validate
Include detection, decision and preparation in the declared recovery timeline.
- 04Restore and validate → Service accepted
An open port or a restored database alone may not satisfy acceptance.
- 05Service accepted → Compare with objectives
Compare the 12-minute interruption with the 15-minute target.
- 06Recoverable point → Compare with objectives
Compare the recovery-point age at interruption with the 5-minute target.
Check your understanding: Is a replica a complete backup strategy?
No. Replication can also copy accidental deletion or corruption. Failover, retained recovery points and a tested restore solve different failure cases. Promotion must prevent conflicting writers and preserve a deliberate failback plan.
Work through it, step by step.
Agree
Define useful service and business tolerance.
What you have now: Separate RTO and RPO targets.
Recover
Restore data into a controlled destination.
What you have now: Actual recovered and missing records.
Validate
Check integrity, compatibility, access and dependencies.
What you have now: Useful-service acceptance.
Reconcile
Resolve missing work and plan any return to the original system.
What you have now: A receiving runbook.
FROM RECOGNITION TO REASONING
Explain the visual, predict a case and compare it with the linked workshop.
- 01
Name each component in plain English.
- 02
Predict the outcome before revealing the explanation.
- 03
Run the linked exercise and retain actual evidence with its limits.
A labeled explanation and a workshop observation you can reproduce.
Compare your reasoning when you are ready
A fast restore can still lose too much work. A replica may copy accidental deletion. Measure the useful path, not only process restart.
Keep notes in your own working copy. Moving through these cards does not record completion.
TURN THE IDEA INTO USEFUL WORK
You can now go deeper.
A fast restore can still lose too much work. A replica may copy accidental deletion. Measure the useful path, not only process restart.
Academy lessons use your existing training access. Free guides stay open.