Palantir Sr Developer with Lead Experience ONLY

Senior Foundry Platform Optimization Analyst Lead
About the Role
We're running a full-platform optimization audit of a production Palantir Foundry environment for a healthcare client — sources and ingest, raw and curated storage, pipelines, ontology, applications and AIP, compute and cost, security and governance, and platform hygiene, end to end. This is a phased, read-only investigation, not a build engagement: no fixes before the final gate, and every finding must carry a specific proposed action or recommended change, not just the observation. The goal is a prioritized, evidence-backed improvement plan — the few things that, if changed, would matter most — not the longest possible list of problems. The Lead analyst we hire onto this engagement owns that outcome end to end, and must be able to independently deliver the complete scope below, start to finish
This is not a checklist exercise. For every step and every domain reviewed, the Lead analyst must go beyond flagging what's wrong and recommend a better, higher-performing design or approach, time and cost saving, the specific change that would improve it, not just the problem itself.
Must Have
· 7+ years on Palantir Foundry in production environments, including at least one engagement where you assessed or audited an existing platform rather than built net-new
· Able to independently investigate and produce evidence-backed findings across the full nine-domain scope:
o Sources, connections & ingest-source inventory, sync cadence vs. demand, agent health, credential and service-account hygiene, duplicate/shadow ingest
o Raw layer & landing datasets-storage footprint, transaction retention, schema drift and stability, abandoned or unread scheduled data
o Transforms & pipelines-Spark internals (stage counts, shuffle volume, task skew), incremental vs. full recompute, redundant logic, test coverage and CI/CD, branch hygiene
o Curated datasets & storage layout-file sizing and partitioning, column bloat, data types and precision, semantic layer consistency
o Ontology-object/link/action type design, write-back patterns, backing dataset bindings, storage generation, change management-strong enough to personally judge whether an ontology reflects the business or just mirrors source tables
o Consumption, applications & AIP-Workshop performance and query patterns, AIP Logic/Agent Studio usage and governance, automation, OSDK/external consumers, export behavior
o Compute, scheduling & cost- Resource Management, Performance Optimization for Long-Running Jobs compute profile tuning, schedule and trigger design, idle/zombie compute, unit economics
o Security, governance & data health- markings and classification, row/column-level restriction, roles and groups, access review cadence, data health checks, audit logging, HIPAA-relevant controls (PHI masking, PHI audit trail)
o Platform hygiene, structure & adoption- naming conventions, ownership and stewardship, documentation coverage, environment/promotion strategy, active adoption
· PySpark and Python at production quality, and comfortable with Foundry pipeline tooling (Pipeline Builder, Code Repositories, CI/CD)
· Comfortable running a phased, read-only investigation with real discipline: no fixes before the gate, every finding tied to a location/measurement/owner, findings confirmed with the owning team before they're written up, severity and confidence scored separately
· Client-facing-comfortable presenting findings to both a technical and an executive audience, including a findings-based leadership readout
· Healthcare data context -Epic FHIR, claims, EHR, or clinical workflows
Nice to Have
· AIP Agent Studio experience-at least one real agent shipped with guardrails
· OSDK + React, for assessing consumption-layer findings first-hand
· Prior consulting, technical due-diligence, or platform-assessment background (inside or outside Palantir)
· Experience standing up a recurring platform review cadence, not just a one-time audit
· SOC 2 / HITRUST compliance audit experience
· Palantir certifications
The Engagement Will Produce
Capable of contributing, at full quality, to all of the following-regardless of which domains they are internally assigned on any given day:
· A machine-readable asset inventory and a 90-day cost/performance baseline
· A finding register scored for severity and confidence, with evidence attached and owners confirmed
· An exception register for material findings outside the standard framework
· The prioritization slate and the leadership readout
· A standards proposal that prevents the same platform debt from reaccumulating

Work arrangement
No