Senior Machine Learning Engineer - Responsible AI
- Compensation
- Not disclosed
- Location
- Bengaluru, IN
- Work setup
- Remote
- Seniority
- Senior
- Employment
- Not specified
- Experience
- 7+ years
- Category
- Ml / AI
- Posted
Skills
Must have
- Bachelor’s or Master’s degree in Computer Science, Machine Learning, Data Science, or a related quantitative field, with 7+ years of industry experience building and deploying machine learning systems
- Deep hands-on experience with modern NLP and LLM techniques, transformer architectures, prompt engineering, fine-tuning (e.g., PEFT/LoRA), and model alignment.
- Proficiency in Python (and familiarity with Kotlin, Java, or Go) with a track record of writing clean, testable, and performant production-grade code.
- Proven experience constructing rigorous evaluation pipelines, benchmarking datasets, and managing precision/recall trade-offs for high-consequence classification or safety systems.
- Expertise at telling stories with data and familiarity with at least one visualization tool (e.g. Tableau, R-Shiny, Microstrategy, SAP Business Objects, Looker, etc.)
- Experience deploying and scaling ML services in cloud environments (AWS / GCP) using containerized platforms (Kubernetes/Docker) and distributed data processing frameworks (Spark, SQL).
- Strong analytical problem-solving skills, ability to navigate ambiguity, and clear communication skills to articulate ML trade-offs to technical and non-technical stakeholders.
- An agile development mindset focused on constant iteration, solving problems from first principles, and navigating ambiguity.
Full description
Overview
Working at Atlassian
Atlassians can choose where they work – whether in an office, from home, or a combination of the two. That way, Atlassians have more control over supporting their family, personal goals, and other priorities. This is a remote position. To help our teams work together effectively, this role requires you to be located in India.
We are seeking a Senior Machine Learning Engineer to join our team, dedicated to developing innovative solutions to make AI services responsible (Responsible AI).
You will join the Responsible AI (RAI) & AI Safety engineering team. We are responsible for building, operating, and advancing the centralized AI safety, content moderation, and guardrail platforms powering Atlassian Intelligence and Rovo.
Our systems run inline and asynchronously across high-throughput production workloads, detecting and mitigating jailbreaks, prompt injection, harmful content, bias, and policy violations while preserving latency and user experience. We partner closely with Product Engineering, Central AI Platform, AI Gateway, Legal, and Trust teams to deliver safety systems that scale globally.
We have teams in the USA, Australia, New Zealand, and this will be our expansion to India. This team is responsible for building solutions and operating mature platforms to improve the entire company's security and trust & safety posture.If you are passionate about improving security, trust & safety, and combating abuse, this role presents a unique opportunity to make a meaningful impact!
Responsibilities
At Atlassian, a Machine Learning Engineer (MLE) is responsible for building, training, evaluating, and deploying machine learning models into production systems at scale. The role sits at the intersection of software engineering and data science, requiring production-grade code to turn research prototypes into real-world systems like search ranking, recommendations, and GenAI platforms.
What You’ll Do
- Design & Productionize Safety Models: Architect, train, fine-tune, and deploy ML/LLM-based guardrail and moderation models (text, multi-modal, and code) to safeguard user inputs and model outputs across Atlassian products.
- Build Robust Evaluation & Benchmarking: Design offline and online evaluation frameworks, adversarial test suites (red-teaming), and LLM-as-a-judge pipelines to measure precision, recall, false-positive rates, and robustness against model drift.
- Optimize for Real-Time Serving: Work at the intersection of model accuracy and low latency (P50/P99 targets) to deliver cost-effective, high-throughput safety filtering integrated with our central AI Gateway and inference layers.