Point your AI agent at freehire and let it find you a job. A CLI and an MCP server over the whole job API — no browser.
Senior Machine Learning Engineer at an AI consultancy: you build LLM-powered systems end-to-end for clients — from rapid proof of concept to production — using Python, RAG, agents, and rigorous evals, deployed to cloud (AWS preferred) via CI/CD, in a fully remote, client-facing LATAM team.
Senior ML engineer optimizing the Kimchi LLM inference stack inside Cast AI — improving throughput, latency, cold starts, and cost per token across a self-run GPU fleet. Day-to-day work spans SGLang/vLLM/TensorRT-LLM engine tuning, KV-cache strategy, weight loading, quantization, and multi-node distributed inference on Python, CUDA, and Kubernetes.
This Role We are seeking a Lead DevOps Engineer to establish and drive our infrastructure automation capability from the ground up. This greenfield opportunity will see you define the frameworks, standards, and tooling…
Principal architect who owns end-to-end architecture of Karman's Digital Environment, a simulation platform for developing, testing, and validating the company's AI power-orchestration systems before production deployment. Deeply hands-on across real-time systems, time synchronization, deterministic replay, and power-systems modeling, while building and leading an engineering team.
Causal Labs is an AI startup building a Large Physics foundation Model to predict and control physical systems, starting with weather. This role builds and optimizes high-throughput, low-latency GPU inference infrastructure (PyTorch/JAX serving, Kubernetes/Ray/Slurm orchestration) so large-scale evaluation and backtesting never gate research.
Software engineer on a newly-formed team building and deploying LLM agents to tactical edge hardware for defense-sector customers: training/fine-tuning edge-capable LLMs, optimizing models for GPU/CPU/accelerator platforms, and interfacing agents with sensors, effectors, and drone systems. Core stack: Rust/Python/C++, PyTorch, HuggingFace, LangChain, vLLM.
Engineer on Alibaba.com (International Station)'s search/ads/recommendation engine team in Hangzhou: builds and optimizes distributed high-concurrency online engines, large-scale TensorFlow/PyTorch model inference services, and LLM Agent applications (tools/MCP, RAG), with focus on latency, throughput, cost, and stability.
Senior Staff ML Engineer defining the architecture of agentic AI systems that turn global trade data (shipments, customs declarations, invoices) into risk, compliance, and enforcement intelligence. Day to day: LLM fine-tuning, RAG and agent pipelines, evaluation/observability frameworks, plus mentoring and technical leadership. Remote or hybrid in the US.
Develops and operates LLM-powered products for the National Bank of Kazakhstan's digital center: wiring tools/context via MCP and APIs, building eval pipelines (DeepEval, Ragas, LLM-as-a-judge), tracing regressions in LangFuse, and deploying via GitLab CI/ArgoCD. Core stack: Python, FastAPI, LangChain/LangGraph, vLLM, Weaviate, Kubernetes.
AI engineer/architect at Alibaba designing the overall architecture of data-annotation and RL environment platforms — sandboxed RL environments, workflow engines, LLM gateways, and multi-tenant security. Requires deep distributed-systems, container/sandbox (Docker, gVisor/Firecracker/Kata), and LLM serving (vLLM/SGLang) expertise.
Builds and maintains automated LLM evaluation pipelines covering general, agent, and persona/role-playing capabilities, running benchmarks (MMLU, HumanEval, BFCL, τ-bench, GAIA) against pre-training checkpoints. Core stack: Python, lm-eval-harness/OpenCompass, and inference frameworks like vLLM and SGLang, on-site in Boon Keng, Singapore.
Senior ML engineer owning optimization of Kimchi's LLM inference stack — improving throughput, latency, cold starts, and cost per token across a self-run GPU fleet. Day-to-day work spans Python services with SGLang, vLLM, TensorRT-LLM, PyTorch, CUDA tooling, KV-cache tuning, quantization, and Kubernetes-based distributed serving.
Senior hands-on AI/ML engineer owning the full lifecycle of LLM and CV/OCR model development — evaluating and fine-tuning open-source LLMs, optimizing inference, and building RAG pipelines for a locally or cloud-hosted AI platform. Core stack: Python, PyTorch, HuggingFace/PEFT/TRL, Unsloth, vLLM, vector DBs, and OpenCV/YOLO/PaddleOCR.
Hands-on ML/GenAI engineering role at India's national payments company: build, train, and deploy models for fraud risk management (streaming inference), AI agents for banking workflows, and fine-tuned SLMs. Core stack: Python, PyTorch/TensorFlow, LangChain/LangGraph, Spark/Flink/Kafka, Kubernetes, MLflow.
A medior cloud ops/platform engineer who owns the cloud operations of DPG Media's GenAI platform: deploying AI services on managed Kubernetes, managing multi-cloud infrastructure (AWS, Azure, GCP) with Terraform, securing secrets, and building CI/CD, while growing into LLMOps.
Algorithm expert on Alibaba Health's AI search team who post-trains and optimizes large language models for search — building SFT/preference data, reward and evaluation systems, and applying SFT, DPO, PPO/GRPO, Rubric/Agentic RL to improve reasoning, tool use, and RAG/agentic search quality in production. Core stack: Python, PyTorch, DeepSpeed/Megatron-LM, vLLM.
Alibaba (Hangzhou) is hiring a system software engineer to build the stack for its AI accelerators (GPU/PPU): kernel and user-mode drivers, a CUDA-compatible runtime, compiler toolchain and PyTorch integration, plus supernode topology management and performance tuning for large-scale training/inference clusters. Core tech: C/C++, Linux, CUDA, PyTorch, vLLM/Megatron.
Job Responsibilities Build and maintain an automated LLM evaluation pipeline covering multiple dimensions, including general capabilities, Agent capabilities, and persona/role-playing. The pipeline should support…
Applied Data Scientist in Binance's Algorithm team building production LLM-powered AI agents for crypto trading — multi-agent systems, reasoning models, and on-chain intelligence. Day to day: designing agent architectures, optimizing intelligence/latency/reliability, and building evaluation frameworks using Python, LLMs, vLLM, and TensorRT-LLM.
A Binance Accelerator Program (internship) data scientist building AI-powered financial trading agents on Web3 data: post-training LLMs with SFT/RLHF so they can use Binance's internal trading skills, maintaining benchmark datasets, and iterating agent workflow prototypes. Core stack: LLM post-training frameworks, PyTorch/TensorFlow, Hugging Face, DeepSpeed, vLLM.
We couldn't check your fit for this role — add a CV to your profile to see it next time.