Forward Deployed Engineer — Data Engineering
A study guide for FDE-flavored Data Engineering roles at enterprise contract-intelligence and document-AI platforms. The Palantir-originated FDE archetype applied to embedded customer deployments — ingestion of messy contract / invoice / PO corpora, SLA-backed extraction accuracy, customer warehouse and ERP integration, and the consultant + engineer duality the role is built on. 17 interactive HTML chapters.
If you only have a few hours
Open 00-START-HERE. Then read 03-fde-archetype and 04-customer-embedding — those framings are tested in every round. Drill 15-interview-questions. Reread 16-day-of the morning of.
Drill: Interview Questions
Hide-show answers · per-Q tracker
Core: Customer Embedding
Discovery · scoping · cadence
Core: SLA-Backed Quality
Extraction eval · HITL · ship-to-spec
Section A · Orient
00
Start Here
Master index, what the loop tests, the single reframe that makes the rest click.
01
The Role, Decoded
FDE — Data Engineering at a contract-intelligence platform: what it actually is, what the JD really means, where you sit between customer and product.
02
Positioning From Scratch
The consultant + engineer duality. How to convey both halves honestly, and how to lean on adjacent experience when one half is thin.
Section B · FDE craft
03
The FDE Archetype
Palantir-originated, now a frontier-AI standard. What makes FDE distinct from a regular SWE or DE, what the round tests for, the three behavioral patterns that disqualify candidates.
04
Customer Embedding
Discovery, scoping, weekly cadence with the customer, the working agreement, kill criteria, escalation paths. The operating model the loop assumes you know.
Section C · Data craft
05
Document Extraction
Patterns for turning unstructured contracts / invoices / POs into structured data — OCR + NLP + CV, layout-aware models, human-in-the-loop, the eval loop that backs an SLA.
06
Pipelines Under Deployment
Ingestion / ETL patterns when the customer's data shape changes weekly, schemas are negotiable, and "production" lives in the customer's cloud. Reprocessing, idempotency, versioned pipelines.
07
SQL & Contract Data Modeling
The contract data model — parties, obligations, renewals, spend rollups, supplier hierarchies. SQL patterns over messy real-world contract data and the joins they imply.
08
Warehouse & Integration
Loading clean output into the customer's Snowflake / BigQuery, integrating with their ERP (SAP, Coupa, Oracle) and BI (Looker, Power BI, Tableau). Identity, auth, refresh cadence, backfills.
09
SLA-Backed Quality & Eval
The differentiating piece. How to measure extraction quality, calibrate human-in-the-loop, ship to a contracted accuracy SLA, and surface failures to the customer before they do.
Section D · Coding
10
Coding Fundamentals
Python patterns FDE-DE rounds reach for — dict-heavy data shaping, generators for streaming, light algos, the "write something I can ship to a customer next week" bar.
11
Coding Problems
Ten drillable problems — extraction-flavored cleaning, schema reshaping, fuzzy entity resolution, reconciliation, evaluation metrics. Multiple approaches per problem.
Section E · Production / Cloud
12
Customer Integrations
The enterprise stack you'll meet — Salesforce, SAP, Coupa, Workday, NetSuite, Snowflake, BigQuery, Looker. API shapes, auth patterns, refresh cadences, the political layer behind each one.
13
Ops & Rollout
Embedded ops hygiene — staging environments at the customer, change management, runbooks, rollbacks, on-call when the customer's data team is your first responder.
Section F · Reference & Execution
14
Domain Context
Contract intelligence and enterprise spend — CLM history, procurement vocabulary (P2P, S2P, S2C), finance / legal personas, where the platform's revenue actually comes from. The vocabulary edge.
15
Practice Interview Questions
~30 drillable Q&A across background, FDE behavior, data craft, customer scenarios, and domain. Hide-show answers, per-question tracker, progress saved locally.
16
Day-Of Tactics
Live-coding moves, customer-scenario recovery patterns, the right questions to ask them, closing statement. Reread morning of.
Study paths
If you have 7+ days
- Day 1: 01, 02 (orient) → 03 (FDE archetype)
- Day 2: 04 (customer embedding), 05 (document extraction)
- Day 3: 06 (pipelines), 07 (SQL & modeling)
- Day 4: 08 (warehouse), 09 (quality & SLA)
- Day 5: 10, 11 (coding drills on a timer)
- Day 6: 12 (integrations), 13 (ops), 14 (domain)
- Day 7: Drill 15. Reread 16. Sleep.
If you have 2–3 days
01, 03 (FDE archetype), 04 (customer embedding), 07 (SQL & modeling), 09 (quality), drill 5 problems from 11, 15 (drill), 16. Skim the rest.
If you have < 24 hours
03 (FDE archetype is the single most leverage chapter), 04 (customer embedding), 09 (quality), 15 (drill all), 16. Skim 07 SQL section. Walk in honest.