AI Research Scientist

ABOUT CLIENT

Our client is a leading research company specializing in technology innovation

JOB DESCRIPTION

Driven by Intellectual Curiosity: Relentlessly explore the unknown, ask sharp questions, and challenge assumptions.
Constructively Combative: Engage in thoughtful debate to uncover truth, refine ideas, and push boundaries.
Experimental Mindset: Design and run AI experiments, analyze outcomes rigorously, and iterate with purpose.
User-Centered Innovation: Adapt AI systems to solve real-world problems and elevate the experience of Menlo's users.
Stay on the Cutting Edge: Continuously learn, absorb state-of-the-art research, and apply it with discernment.
Rapid Prototyping & Iteration: Ideate, implement, debug, and accelerate the cycle of experimentation and insight.
Robust Engineering: Build modular, maintainable training codebases that scale and evolve gracefully.
Efficient Data Infrastructure: Develop high-performance data loaders and utilities to streamline training workflows.
Scalable Training Systems: Architect jobs to run seamlessly across multi-GPU and multi-node setups (e.g., DDP, NCCL).
Performance Optimization: Tune models for speed, stability, and hardware efficiency without compromising accuracy.
Code Quality & Reproducibility: Write clean, testable, and reproducible code that stands the test of time.
Open Source Stewardship: Actively contribute to and improve the broader ecosystem through upstream collaboration.

JOB REQUIREMENT

Strong proficiency in PyTorch, including custom modules, loss functions, and distributed training
Demonstrated experience in training deep learning models in real-world research or production settings
Proficient engineering skills in Python, and optionally C++ for performance-critical components
Experience in working with large datasets, complex pipelines, and real-world debugging
Understanding of training dynamics and the ability to identify and resolve issues
Familiarity with job launchers, logging tools (e.g., Weights & Biases, TensorBoard), and checkpointing systems
Previous involvement with TorchScript, ONNX, or custom inference runtimes
Contributions to PyTorch or open-source ML tooling
Experience with transformer models, diffusion models, or large-scale vision/NLP tasks
Familiarity with batch schedulers (SLURM), cluster environments, and GPU resource management
Ability to collaborate closely with systems engineers or MLOps teams to ensure smooth integration

WHAT'S ON OFFER

Join a top-notch research team in working on impactful projects
Take ownership of and contribute to the development of the core training code infrastructure
Engage with real models, data, and scale, tackling substantial problems
Support in bridging the gap between research speed and engineering excellence
Enjoy a flexible work setup and a culture that promotes depth, clarity, and curiosity

CONTACT

PEGASI – IT Recruitment Consultancy | Email: recruit@pegasi.com.vn | Tel: +84 28 3622 8666
We are PEGASI – IT Recruitment Consultancy in Vietnam. If you are looking for new opportunity for your career path, kindly visit our website www.pegasi.com.vn for your reference. Thank you!

Job Summary

Company Type:

Product

Technical Skills:

Data Science

Location:

Others - Singapore

Working Policy:

Remote

Salary:

Negotiation

Job ID:

J01854

Status:

Close

Related Job:

C++ Engineer - Market Data

Ho Chi Minh, Ha Noi - Viet Nam

Product

  • C/C++

Maintain/enhance our legacy C++-based tick data processing platform as needed, demonstrating product ownership, and helping migrate datasets to our new tick data processing platform. Contribute to our new, modern C++-based tick data processing platform - enhancing the platform to support additional tick data feeds across asset classes, developing/reviewing the implementation of tick data-based interval features/statistics, and adding new functionality to the platform all while maintaining high software standards and best practices. Collaborate with the Research and Portfolio Management organizations to facilitate the transition from our legacy platform to our new platform, supporting their price volume data needs for signal generation.

Negotiation

View details

AI-Native Software Engineering Lead

Ho Chi Minh - Viet Nam

Outsource

  • Backend
  • AI

Responsible for developing and evolving the AI-native SDLC operating model, including agent workflow designs, verification gates, context management standards, and evaluation frameworks Build and lead multi-agent systems using orchestration layers such as Claude Code, GitHub Copilot Workspace, Cursor, LangGraph, CrewAI, or equivalent, from prototype to production Collaborate with the Director of Engineering to contribute to and maintain the company's AI toolchain selection criteria and evaluate tools with engineering rigor, providing internal guidance on when AI is beneficial and when it is not Establish engineering standards, agent evaluation loops, and AI output quality gates across the delivery organization Previous experience in a lead, principal, or staff engineer role with demonstrated cross-team influence Experience in outsourcing, consulting, or multi-client delivery environments Track record of building or leading an internal community of practice, guild, or AI adoption program Develop and continuously evolve the company's AI-native SDLC playbook, including standards, workflow templates, case studies, and guardrails that delivery teams can adopt immediately Design and lead internal upskilling programs that transition engineers from AI-assisted to AI-native working patterns Keep track of the AI capability frontier, model improvements, new agent frameworks, and emerging risks, translating signals into timely updates to client practices Work closely alongside Delivery Teams as an AI transformation advisor and execution partner, identifying the highest-value automation opportunities across the SDLC and coordinating with the team to implement them Design and deploy agent-orchestrated workflows tailored to each client's stack, team maturity, and delivery context, with measurable ROI Build business cases for AI-native adoption with clients and account managers, framing the value in terms of velocity, quality, and cost Represent the company's AI-native engineering capabilities in client conversations, QBRs, and RFP responses as a credible technical authority

Negotiation

View details

Senior System Software Engineer - AI Data Platform - Inference Factory

Ho Chi Minh - Viet Nam

Product

  • Devops
  • C/C++
  • Python
  • Golang

Create infrastructure and tools to automate complex software processes effectively. Improve performance: Deploy advanced test harnesses, benchmarking frameworks, and analytical tools to thoroughly evaluate and enhance the performance and efficiency of software and hardware platforms. Utilize expertise in operating systems, kernel internals, device drivers, memory management, storage, networking, and high-speed interconnects to construct and troubleshoot high-performance systems. Collaborate with engineering teams to comprehend requirements and deliver efficient solutions. Establish performance objectives, assess feedback, analyze data, and continually enhance system reliability. Shape technical strategies: Contribute to developing technical strategies and roadmaps for platform automation initiatives to ensure they are in line with company goals and industry best practices.

Negotiation

View details