CUDA (Compute Unified Device Architecture) Developer

Navtech is looking for a Performance Engineer to characterize, benchmark, and optimize application workloads across CPU and GPU architectures — in bare-metal, virtualized, and distributed/cluster environments. You'll work with cross-functional teams to identify software bottlenecks and drive performance improvements.

Working as a Performance Engineer at Navtech, you will:

• Benchmark and profile workloads on CPU/GPU architectures using tracing tools like VTune and Linux Perf.
• Analyze performance data to identify software bottlenecks and recommend optimizations.
• Tune applications running in cluster and distributed environments.
• Build performance monitoring scripts and automation using C, C++, and Python.
• Apply knowledge of server ecosystem components (memory, storage, networking), virtualization, and Linux internals to performance tuning.
• Document benchmarks and optimization findings for engineering and client stakeholders.
• 
  

Who Are We Looking For Exactly?

• 2+ years of experience in bare-metal and virtual machine environments, with 2+ years of software development in C, C++, and Python.
• Hands-on experience with benchmarking, tracing tools, VTune, and Linux Perf.
• Working knowledge of server architecture, virtualization, and Linux OS internals.
• Experience with distributed/cluster application environments.
• Exposure to simulation technologies (functional, cycle-accurate/timing, performance) and profile-guided optimization is a plus.
• Strong problem-solving, communication, and documentation skills.
• B.S./M.S. in Computer Engineering, Computer Science, or related field.

Qualifications & Requirements:

• 4+ years of experience in C++ application development.
• Hands-on experience with C++11 or above.
• Strong knowledge of object-oriented programming and software design.
• Deep understanding of STL, multi-threading, socket programming, and data structures.
• Solid grasp of Linux development and debugging techniques.
• Proficient in using GCC, GDB, and Makefile.
• Familiarity with Valgrind and similar analysis tools.
• Experience with version control tools like Git.
• Experience writing and maintaining automated tests.
• Experience in capital markets/trading domain is a plus.

Skills:

• Strong problem-solving and analytical thinking.
• Clear and effective communication.
• Self-driven with the ability to work independently.
• Passionate about high-quality software and strong engineering practices.
• Comfortable working in a fast-paced, collaborative environment.

Title                                 : Senior AI Platform / MLOps Engineer

Experience                    : 6+ years

Work type                      : Chennai - Work from Office/other locations - Remote

Employment Type      : Full Time

Notice Period              : Immediate

Work Day                     :Mon to Fri

Key Responsibilities:

• Install, configure and operate OpenShift, NVIDIA GPU operator, OpenShift AI, and NIM microservices on 12× RTX PRO 6000 across two servers; single-node and HA control-plane topologies
• Serving configuration and tuning: quantized model deployment (FP8/FP4), replica balancing, batching, KV-cache and context management
• Azure GPU build environments: provisioning, cost control, parity with the on-prem stack via pinned container/model versions; cloud-to-factory migration with parity regression
• GitOps CI/CD, container registry, artifact/model versioning, environment promotion; observability and audit wiring (Splunk, Prometheus/Grafana)
• Benchmark automation: load harness, p50/p95/p99 latency, tokens/sec, GPU utilization; the capacity report data pipeline
• Platform upgrade procedure with evaluation-regression gates; deployment runbook as a first-class deliverable

Technical Skills:

• 6+ years infrastructure/platform engineering with 3+ years production Kubernetes; OpenShift experience strongly preferred
• Hands-on GPU inference serving in production: NIM, Triton, vLLM, or TensorRT-LLM — you have sized, deployed, and tuned LLM serving on real GPUs and can talk memory-bandwidth trade-offs
• GitOps fluency (ArgoCD/Flux), infrastructure-as-code, container internals; comfortable in air-gapped/proxy-restricted enterprise networks
• Observability depth: metrics, traces, log pipelines; has built performance test harnesses, not just run them

• Azure or AWS GPU compute operations experience

Strongly preferred

• NVIDIA GPU operator and AI Enterprise stack specifics; KServe; Milvus or pgvector operations; VAST/NFS/S3 storage integration; banking or other regulated-environment delivery

About Ampera: 

Ampera Technologies, a purpose driven Digital IT Services with primary focus on supporting our client with their Data, AI / ML, Accessibility and other Digital IT needs. We also ensure that equal opportunities are provided to Persons with Disabilities Talent. Ampera Technologies has its Global Headquarters in Chicago, USA and its Global Delivery Center is based out of Chennai, India. We are actively expanding our Tech Delivery team in Chennai and across India. We offer exciting benefits for our teams, such as 1) Hybrid and Remote work options available, 2) Opportunity to work directly with our Global Enterprise Clients, 3) Opportunity to learn and implement evolving Technologies, 4) Comprehensive healthcare, and 5) Conducive environment for Persons with Disability Talent meeting Physical and Digital Accessibility standards 

Designation: Lead MLOps Engineer (GPU Optimization)

About CloudKeeper: 

CloudKeeper is a cloud cost optimization partner that combines the power of group buying & commitments management, expert cloud consulting & support, and an enhanced visibility & analytics platform to reduce cloud cost & help businesses maximize the value from AWS, Microsoft Azure, & Google Cloud. 

A certified AWS Premier Partner, Azure Technology Consulting Partner, Google Cloud Partner, and FinOps Foundation Premier Member, CloudKeeper has helped 350+ global companies save an average of 20% on their cloud bills, modernize their cloud set-up and maximize value — all while maintaining flexibility and avoiding any long-term commitments or cost. 

CloudKeeper hived off from TO THE NEW, a digital technology service company with 2500+ employees and an 8-time GPTW winner. 

To know more, please visit - https://www.cloudkeeper.com/ 

Responsibilities:

  - Drive R&D and engineering for AI Infrastructure optimization within CloudKeeper's FinOps for AI platform — building the Tuner AI / Commit AI capability on GPU and ML workloads

  - Design and build optimization engines for GPU right-sizing, idle shutdown, spot migration with checkpoint/resume automation, inference batching, quantization, and model placement

  - Extend the optimization stack to LLM-era workloads — caching, model routing, dynamic batching, prompt optimization, RAG-aware architectures

  - Partner with the Lens AI team to translate GPU and ML workload signals into actionable, dollar-quantified optimization recommendations for customers

  - Work cross-functionally with product, platform, and customer success teams to ship optimization features end-to-end (data ingestion → optimization engine → customer-facing recommendation)

  - Lead technical direction for AI workload optimization, set engineering standards, and mentor the ML / MLOps engineering bench as the AI Infrastructure pillar scales

  - (Lead level) Hire, ramp, and grow a team of ML infrastructure engineers as headcount expands

  Must Have:

  - B.E / B.Tech / M.Tech / MCA with 7+ years of hands-on engineering experience

  - Production experience with GPU workloads — has measurably optimized GPU utilization,  throughput, or cost in a real production environment, not just academic / lab work

  - Strong performance engineering background — must come ready with a concrete optimization story including before/after metrics (latency, throughput, or cost reduction)

  - Strong Python + Linux + systems fundamentals

  - Solid understanding of the ML model lifecycle — training, serving, inference — able to reason about what is running on the GPU and why

  - MLOps fluency — model deployment, monitoring, observability, GPU cluster operations

  - Hands-on with cloud GPU instances (AWS P5 / G6, Azure ND series, GCP A3, or equivalent) and Kubernetes-based GPU orchestration (EKS / AKS / GKE GPU node pools, Karpenter, Run:ai, NVIDIA GPU Operator, or similar)

  - Familiarity with at least one modern LLM inference framework — vLLM, TGI, Triton, SGLang, Ray Serve, or BentoML

  - Strong communication skills — able to translate deep technical optimization into customer / business outcomes

  - (Lead level) Experience managing or technically leading a team of 3+ engineers

  Good to Have:

  - Deep LLM-era optimization expertise — KV caching, semantic caching, model routing, dynamic batching, quantization (FP16 → INT8 → INT4), model distillation, structured outputs

  - Familiarity with LLM workload patterns — RAG, agents, embeddings, vector databases (Pinecone, Weaviate, Qdrant)

  - CUDA, NCCL, mixed-precision training and inference

  - Experience with managed ML training platforms — SageMaker, Azure ML, Vertex AI, Databricks Mosaic

  - Exposure to GPU-native clouds — CoreWeave, Lambda Labs, RunPod, Crusoe

  - Open source contributions to ML infrastructure projects — vLLM, llama.cpp, TGI, Ray, Triton, KubeRay

  - Adjacent experience in cloud cost optimization / FinOps — Spot.io, ScaleOps, Granulate, CAST AI

  - Comfort with Agile methodology and modern engineering practices (CI/CD, code review, observability)

Position Title: Real-Time Computer Vision & Edge AI Engineer (Founding Engineering Team / Core LLD) 

Reporting Structure: High-Level AI Architect (Principal ML Scientist, Google) 

Domain: Sub-16ms Edge AI, 3D Pose & Shape Estimation (SMPL-X), TensorRT C++ Inference, Zero-Copy Systems 

Performance Benchmark: Hard locked 60 FPS (<16.6 ms total frame budget) on dedicated RTX hardware

1. Position Overview & Architecture 

We are building a proprietary, ultra-low-latency spatial computing platform centered on high-fidelity 100% 3D Digital Twin architecture and real-time human digitization. 

In this role, you will serve as the Low-Level Design (LLD) Core AI Engineer, working directly alongside a Lead AI Scientist from Google. Your primary mandate is to solve complex surface occlusion and volumetric estimation challenges by building an ultra-fast C++ inference pipeline. This system must accurately regress a subject's true underlying 3D body shape and skeletal pose directly from a live camera feed. You will deploy models that extract parametric data (SMPL-X shape/pose parameters) and bridge these joint rotations seamlessly into our Vulkan graphics engine via shared GPU memory.

System Architecture: 

● Hardware Camera Ingestion: (V4L2 / GStreamer / CUDA) 

↓ Raw RGB Frames (Zero CPU Copy) 

● Edge AI Inference: (TensorRT / ONNX C++ API for 3D Pose Tracking, Kinematic Anchoring, SMPL-X Shape) 

↓ 3D Skeletal Transforms & Shape Parameters 

● Zero-Copy Shared Memory: (CUDA-Vulkan Bridge feeding directly into OpenRigLogic / MetaHuman Engine) 

2. Key Responsibilities & Deliverables 

A. Real-Time 3D Pose & Shape Estimation 

● Deploy and optimize state-of-the-art 3D human body reconstruction models (e.g., Shapy, SMPLify-X, CLIFF) to accurately regress the user's underlying skeletal structure and body volume, effectively bypassing unpredictable surface topologies and complex environmental occlusions. 

● Extract mathematically stable shape parameters (β) and pose parameters (θ) to drive the skeletal hierarchy of a high-fidelity digital avatar. 

B. Edge Inference Pipeline (TensorRT) 

● Translate Python-based research models into production-grade C++ inference engines using NVIDIA TensorRT and ONNX Runtime. 

● Implement INT8/FP16 quantization, layer fusion, and custom CUDA plugins to ensure the entire AI inference pass executes within a strict <10 ms budget per frame. 

C. Temporal Smoothing & Anti-Jitter Kinematics

● Implement highly optimized temporal filters (Kalman filters, One-Euro filters, optical flow tracking) in native C++ to eliminate all high-frequency jitter from the output joint rotations before they reach the graphics engine. 

● Ensure kinematic constraints (e.g., fixed bone lengths) are strictly maintained to prevent the digital asset from stretching or warping dynamically. 

D. Zero-Copy Ingestion & Engine Synchronization 

● Build hardware-accelerated video capture pipelines using V4L2 or GStreamer to ingest raw camera frames directly into GPU memory. 

● Bridge the output coordinate data and transformation matrices to the graphics team using POSIX shared memory and CUDA-Vulkan interop (VK_KHR_external_memory_fd), eliminating CPU staging overhead. 

3. Technical Qualifications & Tech Stack 

● Core Programming: Production-level Modern C++ (C++17/20), Python (strictly for model training/validation), and CUDA C/C++. 

● AI & Acceleration Frameworks: NVIDIA TensorRT, ONNX Runtime (C++ API), PyTorch. 

● Computer Vision Libraries: OpenCV (CUDA backend), MediaPipe C++ bindings. 

● Mathematical Foundations: 3D Kinematics, Matrix Transformations, Quaternions/Euler angles, statistical body modeling (SMPL/SMPL-X architecture). 

● Systems Architecture: Low-latency memory management, multi-threading (std::jthread, lock-free queues), SIMD vectorization. 

4. Relevant Projects & Demonstrable Experience (Preferred) 

Candidates will be preferred if they present functional codebases, GitHub repositories, or thesis work covering:

● Real-Time Body Fitting / Pose Estimation: Practical experience deploying 3D human pose or shape reconstruction models on live video feeds. 

● TensorRT / C++ Deployment: Demonstrable experience stripping a PyTorch model out of Python and running it natively in C++ using TensorRT or ONNX, ideally with custom CUDA layers or INT8 calibration. 

● High-Throughput Vision Pipelines: Built a C++ video processing pipeline that aggressively minimizes latency and avoids memory garbage collection pauses. 

● Kinematics & Smoothing: Applied mathematical filters to raw sensor or AI data to produce smooth, mechanically accurate 3D rotations. 

5. Compensation & Engagement Structure 

● Compensation: ₹1,50,000 to ₹2,00,000/month 

● Mentorship: Direct architectural guidance, algorithm review, and technical leadership from a Principal ML Scientist at Google. 

● Hardware: Dedicated high-end workstation equipped with discrete NVIDIA RTX hardware.

Dear Candidate,

We are currently hiring for the position of C++ / C# Windows Desktop Application Developer for one of our esteemed clients.

Job Description

We are looking for an experienced Windows Desktop Application Developer with expertise in C++ and C# to design, develop, maintain, and enhance Windows desktop applications for hardware products.

The candidate should have hands-on experience in Windows application development, Windows operating system concepts, Win32 API, Windows SDKs, and Visual Studio. Experience in the laptop and desktop domain is preferred.

Key Responsibilities

• Design, develop, and maintain Windows desktop applications using C++ and C#.
• Develop and enhance Windows applications using Windows programming technologies.
• Work with Win32 APIs, Windows SDKs, and relevant libraries and frameworks.
• Analyze, troubleshoot, and resolve application defects and technical issues.
• Migrate and maintain code using Visual Studio IDE.
• Perform unit testing for newly developed or modified code.
• Work with Windows registry, certificates, file and folder security, and Windows OS functionalities.
• Participate in code reviews, technical discussions, and Agile development activities.
• Collaborate with cross-functional teams to improve product functionality and application performance.

Required Skills

• Hands-on experience in C++ – Mandatory
• Hands-on experience in C# – Mandatory
• Minimum 1 year of experience in both C++ and C# is required.
• Experience in Windows desktop application development.
• Knowledge of Windows operating system concepts.
• Experience with Win32 API and Windows SDKs.
• Familiarity with Visual Studio and software debugging.
• Understanding of unit testing, troubleshooting, and defect resolution.
• Experience in the laptop and desktop hardware domain is preferred.

Opportunity Details

Role: C++ / C# Windows Desktop Application Developer

Experience: 4+ Years

Location: HP Office, Bangalore

Work Mode: Work from Office

Mandatory Skills

C++ | C# | Windows Desktop Application Development | Windows OS | Visual Studio

Key Skills

C++ | C# | Windows Applications | Win32 API | Windows SDK | Visual Studio | Debugging | Unit Testing | Windows Registry | Windows OS

• Senior Machine Learning Engineer – Vertex AI
• Senior Generative AI Engineer – GCP
• Google Cloud AI/ML Engineer
• MLOps Engineer – Vertex AI
• LLM Engineer – RAG, Gemini, and Vertex AI