Senior Backend Engineer, Clean-Room Query Engine
placino is a data clean room for regulated enterprises. We let multiple parties run privacy-preserving SQL and ML across each other's data, without exposing raw records. Each party contributes only encrypted, purpose-scoped columns; only privacy-preserving query results cross the boundary, gated by differential privacy, k-anonymity thresholds, and per-query policy enforcement.placino is delivered as SaaS. We operate the whole platform, and every tenant gets its own isolated environment with its data encrypted under that tenant's own key. Kubernetes-native. Terraform-managed. Infrastructure-as-code from day one.We are 11–50 people, headquartered in Dubai (IFZA Business Park, Dubai Silicon Oasis), serving regulated enterprises across finance, health, telco, and AdTech.
The role
You will work on the clean-room query engine, the core IP. Where SQL meets MPC and differential privacy. Where every query has to produce both a useful answer and a defensible audit record.This is a senior IC role focused on correctness, performance, and cryptographic discipline.
What you will work on
- The clean-room SQL execution engine: query planning and aggregation across N parties' contributed datasets, including cost-based strategy selection (direct, split-merge, hash join).
- Differential privacy primitives: noise calibration, privacy budget enforcement, query rejection when a result would leak past ε.
- K-anonymity and l-diversity threshold enforcement at result-aggregation boundaries, including edge cases auditors actually probe.
- MPC integration (secure aggregation, secret sharing) for queries where DP noise alone would not be sufficient.
- Privacy-preserving ML primitives: lookalike modeling and DP-bounded model outputs.
- Audit trail subsystem: every query produces cryptographic evidence mapped to GDPR Art. 32, HIPAA Privacy Rule, and ISO 27001 controls.
You are a fit if
- 6+ years building production distributed systems in Go, Rust, C++, or modern Java/Scala.
- Hands-on with database internals OR distributed query engines (Trino, Spark, ClickHouse, DuckDB).
- Familiarity with privacy-enhancing technology: differential privacy, secure multi-party computation, zero-knowledge proofs, or homomorphic encryption (at least two of these in working depth).
- Strong understanding of distributed consensus, eventual consistency, isolation levels, and the failure modes that come with them.
- Pragmatic about correctness: you write proofs (or proof sketches) in code reviews, not just unit tests.
Bonus
- Open-source contributions to Trino/Presto, Apache Arrow, OpenDP, OpenMined PySyft, or similar.
- Background in academic privacy research or applied cryptography.
- Experience with confidential computing runtimes: Intel SGX, AMD SEV, AWS Nitro Enclaves.
- Published or contributed to PETs standards work (NIST, ISO, IETF).
- Familiarity with formal verification tools (TLA+, Coq, Dafny), even at the \"I have used it on a side project\" level.
What we offer
- Competitive compensation aligned to senior IC market rates.
- Equity participation.
- Hybrid Dubai HQ + remote-first culture; GCC + EMEA collaboration windows.
- Healthcare, learning budget, conference attendance.
- Real ownership: the clean-room query engine is the company. What you ship is what customers buy.