SDE II - AI Platform

Job description

Most boards and executives are currently flying blind when it comes to cyber risk. They are guessing. At Safe, we’ve built an AI-driven engine that finally gives the C-Suite a clear, quantified, and real-time view of their security posture. We don’t just provide data; we provide certainty.
We are a $170M Series C-funded category leader. We don’t play in the mid-market; we operate at the highest levels of global enterprise. Today, we are proud to serve 10% of the Fortune 500, protecting global icons such as Apple, Netflix, AT&T, Verizon, and Victoria’s Secret.
As we scale toward our next chapter, we are looking for high-performers who want to do the best work of their careers at the intersection of AI and Cybersecurity.
The Culture Memo: Our Operating System
Safe is not a typical corporate environment. We are a high-intensity, mission-driven team. We value builders who want to define a category and work alongside people who are equally committed to excellence.
Extreme Ownership: We don’t do "not my job." We hire people who see a gap and own the solution from start to finish.
The Elite Standard: We serve the most sophisticated companies on the planet. Our work must be bulletproof. Whether it’s a line of code or a sales deck, we aim for Tier-1 quality every time.
Methodology & Rigor: We don’t wing it. From Force Management and MEDDICC in sales to data-driven sprints in engineering, we rely on proven frameworks to stay disciplined and predictable.
Radical Candor: We move too fast for politics or sugar-coating. We value direct, honest feedback that helps us find the right answer quickly.
The Series C Hustle: We have the stability of a well-funded leader but the heart of a startup.
The Perks & Ownership:
We want our team to feel like owners because they are owners. We trust our people to manage their results and their time.
Meaningful Equity: Every "Safestar" is a shareholder. You aren’t just an employee; you are a partner in our success.
Unlimited Leaves: We don’t believe in clock-watching. We offer unlimited leave because we trust you to take the time you need to recharge while staying committed to the mission.
Comprehensive Benefits: We provide top-tier medical insurance and wellness benefits to ensure you and your family are well cared for.
Career Trajectory: We are growing aggressively. For high-performers, the path for advancement moves at the speed of your ambition.

AI is not a side project at Safe - it is the engine behind how we quantify cyber risk for the world's largest enterprises. Agentic services, LLM analytics, and inference workloads run in production, on real customer data, under enterprise security constraints.
 
You will own the platform underneath all of it: GPU clusters, inference serving, model lifecycle, and the developer-facing abstractions on top, so every engineering and AI/ML team at Safe ships AI workloads without rebuilding infrastructure each time.
 
This is a platform and infrastructure engineering role, not an ML research role. We want an engineer who has operated GPUs in production, not only consumed them.

AI platform is your primary charter. As an SDE II on the platform team, you will also contribute to core platform engineering, multi-tenant microservices, APIs, and cloud architecture, and carry the same review, mentoring, and delivery ownership as every SDE II at Safe.

What You'll Do:
•
Build the AI platform: APIs, SDKs, and self-service workflows so AI/ML engineers deploy, version, evaluate, and monitor workloads without touching raw infrastructure.

•
Own GPU infrastructure: Cluster design, provisioning, scheduling, isolation, and utilisation optimisation across shared multi-team demand.

•
Run inference in production: vLLM, Triton, KServe, or Ray — continuous batching, autoscaling, model routing, and multi-model serving against real latency and throughput SLOs.

•
Optimize relentlessly: Drive tokens/sec, p99 latency, GPU utilization, and cost per token through quantization, batching strategy, and capacity planning.

•
Engineer GPU-aware Kubernetes: GPU Operator, device plugins, GPU-aware scheduling, MIG partitioning, and distributed workloads over NCCL.

•
Debug the hard layer: GPU OOMs, driver and CUDA runtime mismatches, interconnect bottlenecks, throughput regressions.

•
Build the MLOps backbone: Model registry and versioning, CI/CD for AI workloads, and safe rollout/rollback.

•
Own the evaluation layer: Offline and online eval harnesses, golden datasets, LLM-as-a-judge scoring, accuracy and hallucination metrics, and automated regression gates so no model, prompt, or quantisation change ships without a measured quality verdict.

•
Make it observable: Response quality and accuracy drift, latency, throughput, GPU utilisation, and per-tenant cost — with SLOs, alerting, and incident response.

•
Secure and multi-tenant by default: AuthN/AuthZ, secrets, data protection, tenant isolation, and governed access to models and GPUs.

•
Codify and lead: Terraform f

Prepare for the interview

Nothing collected for this employer yet. The Blind 75 is what technical screens draw from; practise it here, with a coach, in Java or Python.

More at Safe Security

• Vulnerability Researcher II · New Delhi
• SDE III - Engineering Productivity (AI) · Bengaluru
• Staff Engineer · Bengaluru
• Customer Success Advisor · Remote, Americas