Lead Test Engineer
LOCATION: GREATER BENGALURU AREA
Company Description
We are looking for exceptional talent and leadership to join Fast Growing Startup into Scalable Intelligence, the world’s first company developing Agentic Silicon for powering the future of AI.
Founded in 2023, We have deep customer engagements across America, Europe, and Asia, and demonstrated functional prototypes to prove our concept and vision.
Location: Bangalore, Karnataka
At
Tsavorite Scalable Intelligence
, we are pioneering the semiconductor industry’s first
Omni Processing Unit (OPU)
—a breakthrough composable architecture designed to power the next generation of real-time, multi-modal
Agentic AI
.
Founded in 2023 by a veteran team from
Intel, Nvidia, Qualcomm, and Apple
, we are on a mission to eliminate the cost, power, and complexity bottlenecks of legacy GPU systems. Our technology delivers a 10x performance gain at 10% of energy consumption, scaling seamlessly from edge devices to exascale data centers.
Why Engineers Join Us:
- Architectural Innovation:
Work on the
MultiPlexus™ fabric
, a revolutionary interconnect offering petabyte-scale bandwidth and ultra-low latency.
- Full-Stack Impact:
We are building everything from modular chiplets on
Samsung’s SF4X platform
to our
Tsavorite AI Orchestration Stack (TAOS)
, which provides zero-switching-cost compatibility for CUDA-optimized workflows.
- Market Momentum:
We emerged from stealth with over
from Global 500 companies and sovereign cloud providers.
Key Links to Include
- Official Website:
tsavoritesi.com
- LinkedIn Page:
Tsavorite Scalable Intelligence | LinkedIn
- Recent News:
Tsavorite Emerges with $100M in Orders & 10x Performance (Forbes)
- https://www.businesswire.com/news/home/20251110678526/en/Tsavorite-Scalable-Intelligence
Role : Lead Test Engineer
Role Overview
We are seeking a Lead Test Engineer to bridge the gap between high-level AI frameworks and low-level hardware execution. You will lead the quality engineering effort for our compiler stack, ensuring that models originating in PyTorch, ONNX are correctly and efficiently lowered through torch-mlir, Triton, and LLVM.
The ideal candidate is a systems-thinking engineer who can balance the rigor of LLVM lit for IR-level verification with the flexibility of pytest for high-level functional and performance validation.
Core Responsibilities
* Hybrid Testing Strategy: Design a dual-layered testing strategy:
* Low-Level: Use LLVM lit like framework for fine-grained verification of MLIR transformations, optimization passes, and IR correctness.
* High-Level: Build a comprehensive pytest suite to validate end-to-end model accuracy, operator coverage, and integration between PyTorch and the compiler backend.
* Triton & Kernel Testing: Develop robust test harnesses for Triton kernels.
* Performance Benchmarking & Analysis: * Architect automated pipelines to track kernels/ compiler generated code latency, memory bandwidth, and compute utilization.
* Use profiling tools to root-cause performance regressions and hardware bottlenecks.
* Create data-driven performance "gates" to ensure every release meets strict efficiency KPIs.
* Compiler Lowering Validation: Own the verification of the compiler stack.
* Automation & CI/CD: Lead the integration of these testing layers into a seamless CI/CD pipeline.
Required Technical Skills
* Testing Tooling: Proficiency in pytest (advanced fixtures, parametrization, plugins) and LLVM lit.
* ML Ecosystem: Experience with PyTorch (FX, Dynamo, or TorchScript), DSLs, etc.
* Compiler Internals: Good understanding of LLVM and MLIR infrastructure.
* Programming: Mastery of Python for high-level testing and C++ for systems-level verification.
* Performance Mastery: Proven ability to profile kernels and analyze hardware-level execution metrics (stalls, cache hits, occupancy).
* Hardware: Hands-on experience with Triton and a good understanding of GPU or specialized AI accelerator architectures.
Education & Experience:
* Bachelor’s or Master’s degree in Computer Science, Computer Engineering, or a related technical field.
* Experience: 5+ years in systems software testing, compiler development, or high-performance computing (HPC).
Preferred Qualifications
* Bachelors / Masters in the relevant field
* Experience with ML software stack ecosystem
* Demonstrated leadership in establishing QA standards for complex systems software.
Contact
Uday
muday_bhaskar@yahoo.com
www.mulyatech.com
"Mining the Knowledge Community"
