Flink Engineer — Open-Source Streaming Platforms

Location: Remote, based in India

Department: Engineering — Open Source Streaming Platforms

Reports to: CTO / VP Engineering

About Deep.BI

Deep.BI provides enterprise support, consulting, and engineering expertise for open-source analytics and data infrastructure platforms such as Apache Flink, Apache Druid, Apache Kafka, StarRocks, Apache Cassandra, and other emerging technologies.

Our customers run high-volume, mission-critical streaming and analytical systems. They rely on us to help them design better architectures, build better pipelines, improve performance, choose the right technologies, and use open-source systems correctly at scale.

We are a small, senior, remote-first team working across the US, Brazil, Europe, India, and the Philippines. We support 100+ customer environments, from advisory consulting and architecture reviews to advanced engineering support and 24/7 incident coverage.

Role Overview

As an Apache Flink Engineer, you will be one of Deep.BI's core technical experts for streaming data systems, with a primary focus on Apache Flink application design, development best practices, internals, performance, correctness, and architecture.

We are looking for someone who deeply understands how Flink works and how real-world Flink applications should be designed, built, reviewed, optimized, upgraded, and evolved.

You should be comfortable discussing topics such as DataStream API, Table/SQL API, stateful stream processing, event time, watermarks, exactly-once semantics, checkpointing, savepoints, connectors, serialization, schema evolution, job upgrades, backpressure, performance tuning, and Flink deployment patterns.

You will help customers build, review, debug, optimize, and evolve Flink-based systems. This includes reviewing application code and architecture, helping with pipeline correctness, improving performance, supporting migrations, troubleshooting production issues, and joining urgent customer escalations when deep Flink expertise is needed.

The ideal candidate is a strong Flink engineer with deep technical judgment, strong customer-facing communication skills, and enough production experience to support serious enterprise environments.

Key Responsibilities

Flink Engineering Expertise

Help customers design, build, and improve Apache Flink applications.
Review Flink job architecture, dataflow design, state usage, operator structure, serialization, and connector choices.
Advise on DataStream API, Table API, Flink SQL, CEP, Process Functions, timers, stateful operators, and custom connectors.
Guide customers on event-time processing, watermarks, late data handling, windowing, joins, deduplication, enrichment, and exactly-once design.
Help customers avoid common Flink development mistakes that lead to correctness, latency, or scalability problems.
Support Flink version upgrades, API changes, savepoint compatibility, schema evolution, and migration planning.
Evaluate whether Flink is the right tool for specific customer use cases, and help position it correctly against Kafka Streams, Spark Structured Streaming, Beam, NiFi, or custom services.

Architecture & Design Support

Design or review end-to-end streaming architectures involving Flink, Kafka, StarRocks, Druid, Iceberg, Paimon, object storage, and analytical databases.
Help customers choose between Flink deployment models, including Kubernetes, Flink Operator, standalone clusters, cloud-managed services, and self-hosted environments.
Advise on high availability, disaster recovery, state management, replay strategy, and operational safety.
Prepare technical recommendations, architecture notes, design reviews, migration plans, and implementation guidance.
Work with customers who are adopting Flink for the first time, migrating from managed services, or replacing legacy streaming systems.

Flink Internals, Performance & Correctness

Diagnose complex behavior related to state backends, checkpointing, savepoints, timers, watermarks, backpressure, operator chaining, network buffers, and job graph execution.
Analyze performance issues in Flink jobs, including latency, throughput, checkpoint duration, RocksDB behavior, Kafka lag, skew, serialization overhead, and inefficient operator design.
Help tune Flink applications and clusters for low latency, high throughput, and predictable recovery.
Review metrics, logs, flame graphs, thread dumps, heap dumps, JVM behavior, and resource usage when needed.
Identify whether issues are caused by application logic, Flink configuration, connector behavior, cluster resources, Kafka behavior, downstream sinks, or Flink bugs.

Customer-Facing Technical Communication

Explain complex Flink concepts clearly to engineers, architects, and technical leaders.
Write technical reports, architecture reviews, design recommendations, tuning notes, and customer-facing explanations.
Join technical workshops with customers to review Flink pipelines, discuss implementation choices, and propose improvements.
Help Deep.BI produce high-quality technical material related to Flink, including blog posts, internal playbooks, training content, and reusable consulting assets.

Open-Source & Engineering Contribution

Track important changes in Apache Flink releases, FLIPs, connector updates, and ecosystem developments.
Contribute to internal tools, examples, diagnostic scripts, reference architectures, and best-practice templates.
Contribute upstream to Apache Flink, connectors, documentation, or related projects when possible.
Help Deep.BI strengthen its position as a leading independent expert in open-source streaming technologies.

Production Support & Incident Participation

Participate in advanced troubleshooting for customer production environments.
Join urgent customer escalations when deep Flink expertise is needed.
Participate in a paging / on-call rotation for P0/P1 incidents as part of Deep.BI's enterprise support offering.
Help with root-cause analysis, prevention recommendations, and long-term fixes after major incidents.

Requirements

3+ years of hands-on experience with Apache Flink in real-world production, platform, or serious engineering environments.
Strong understanding of Flink application development.
Deep knowledge of Flink DataStream API and/or Flink SQL / Table API.
Strong understanding of stateful stream processing, keyed state, operator state, timers, checkpoints, savepoints, and recovery.
Practical experience with event time, watermarks, windowing, late data, joins, deduplication, and streaming correctness.
Experience designing or reviewing Flink jobs that interact with Kafka and analytical stores or databases.
Strong knowledge of Flink performance tuning, including backpressure, checkpointing, state backend behavior, parallelism, resource allocation, and serialization.
Familiarity with Flink deployment on Kubernetes, Flink Operator, or similar production environments.
Solid understanding of distributed systems concepts: fault tolerance, consistency, partitioning, replay, ordering, exactly-once semantics, and failure recovery.
Good working knowledge of Kafka as a Flink source/sink.
Comfort with Linux, JVM basics, SQL, logs, metrics, and debugging distributed systems.
Strong English communication skills, especially written technical communication.
Based in India and able to work India / APAC business hours with regular overlap into European hours.
Willingness to participate in a paging / on-call rotation for urgent P0/P1 incidents.

Strong Plus

Experience contributing to Apache Flink, Flink connectors, Apache Beam, Kafka, Iceberg, Paimon, or related open-source projects.
Experience building custom Flink connectors, source/sink integrations, or complex stateful operators.
Experience with Flink CDC, Debezium, Iceberg, Paimon, Delta Lake, Hudi, StarRocks, Druid, ClickHouse, Cassandra, or similar systems.
Experience migrating workloads from Spark Streaming, Kafka Streams, Kinesis Data Analytics, Confluent Cloud, Databricks Streaming, or custom streaming platforms to Flink.
Experience with Flink upgrades, savepoint migrations, schema evolution, and compatibility issues.
Experience with Java or Scala at a strong engineering level.
Experience with Python/PyFlink is useful but not sufficient by itself.
Experience providing external technical consulting or enterprise support to customers.
Experience with Kubernetes, Helm, Terraform, cloud infrastructure, or DevOps practices.
Experience writing technical blogs, documentation, training materials, or conference talks.

What We Offer

Work with advanced open-source streaming and analytics technologies at enterprise scale.
Solve difficult engineering problems across Flink, Kafka, Druid, StarRocks, and related systems.
Collaborate with senior engineers and real-world users of open-source data infrastructure.
Opportunity to contribute to open source, technical content, and public thought leadership.
Competitive compensation.
Support for conference travel and OSS contributions.
Flexible contract structure: direct employment, global payroll partner, or B2B contractor, depending on preference and legal setup.

Location & Working Style

Remote-first, based in India.
Regular overlap with European business hours is required.
Occasional overlap with North American hours may be needed for customer calls, handoffs, or major escalations.
Participation in a paging / on-call rotation for urgent P0/P1 incidents is required.

Apache Flink Engineer / Flink Expert