Member of Technical Staff (Model Quantization & Inference)

About Lamb Labs

At Lamb Labs, we're building a new AI chip: the Model Processing Unit (MPU).

Modern AI inference is constrained by the cost of moving model weights between memory and compute. We're taking a fundamentally different approach: aggressively compressing and optimizing language models so that their weights and computation can be encoded directly into silicon.

We're developing extremely low-bit language models and custom hardware architectures together, with the goal of making inference orders of magnitude faster and more energy-efficient.

We're a small, research-driven team working at the intersection of machine learning, compilers and computer architecture. We care about fundamental ideas, rigorous experimentation and turning research into working systems.

The Role

We're looking for an exceptional ML engineer or researcher to work on some of the hardest problems in low-bit language model inference.

You'll help develop the models and algorithms that make our hardware architecture possible.

Our focus is on pushing language models to extremely low precision, particularly 1–2-bit weights, while preserving as much model quality as possible. But this goes well beyond weight quantization.

You'll work across activation quantization, KV-cache optimization, attention, numerical representations, training and fine-tuning, and the broader challenges of running highly compressed language models efficiently.

We're not looking for someone who simply knows how to use existing quantization libraries. We want someone who understands these methods deeply, can identify their limitations, and can develop better ones.

You'll have broad ownership of the problem, from understanding the mathematical and numerical behavior of models to implementing new techniques, running experiments and working directly with our hardware team.

At Lamb Labs, the model and the chip are designed together. Your research will directly influence how we build our silicon.

What You'll Work On

Extreme low-bit quantization

Develop and improve methods for quantizing large language models to sub-4-bit precision, particularly 1–2 bits per weight. Investigate binary, ternary, mixed-precision and other hardware-efficient representations. Understand and overcome the quality degradation caused by aggressive quantization.

Activations, attention and KV-cache

Explore low-bit activation quantization, including outlier handling, scaling, dynamic ranges and numerical stability. Develop methods for reducing KV-cache memory requirements while preserving attention quality and long-context capabilities. Investigate alternative attention implementations and model representations that are more efficient in hardware.

Training, fine-tuning and model quality

Work across post-training quantization (PTQ), quantization-aware training (QAT), fine-tuning, distillation and retraining where appropriate. Develop methods to recover model quality after extreme compression. Understand how architectural choices, training procedures and quantization interact.

Hardware-aware model optimization

Collaborate closely with our FPGA and ASIC engineers to develop models and algorithms that map efficiently onto custom silicon. Explore trade-offs between model quality, bit width, computational complexity, storage and hardware efficiency. Help determine which operations and numerical representations should be supported directly in hardware.

Inference research and engineering

Build and improve inference implementations for highly compressed language models. Understand and optimize transformer inference, including prefill, decode, attention and KV-cache behavior. Design rigorous experiments and evaluations to measure model quality, latency, throughput and resource requirements.

You'll have significant freedom to pursue the approaches that work, whether that means developing new quantization algorithms, modifying model architecture