Want to build systems that actually hold up under long-running AI workloads?

Most agentic systems for science don’t fail at the model layer. They fail because the infrastructure can’t support long-horizon execution.

You’ll join a team building autonomous AI agents that run full research cycles. Ingesting thousands of papers, forming hypotheses, running experiments, and producing traceable outputs used by real scientific teams.

The challenge is making that work in production.

You’ll own the systems behind it. APIs, data pipelines, and platform architecture designed for long-running workloads, large-scale ingestion, and iterative experimentation loops. This is full-stack in scope, but backend in depth, where system design decisions directly impact what the platform can do.

You’ll be working across:

• Backend services in Python or Node, building scalable APIs (FastAPI/REST)
• Data pipelines supporting agent execution and scientific workflows
• Cloud infrastructure (AWS/GCP), containerisation (Docker, Kubernetes)
• CI/CD, observability, and reliability for systems under continuous load

This isn’t a generalist full-stack role. You’ll need to understand how systems behave under heavy data and compute demands, and be comfortable making architectural trade-offs across distributed systems.

The team is small, high-calibre, and already running real workloads with revenue traction. Backed by $70M+, they’re building infrastructure that defines how AI is applied to scientific discovery.

Salary: $200,000–$350,000 + equity
Location: San Francisco (onsite)

Senior Applied Researcher

Want to build vision-language models that understand complex, real-world environments?

You’ll join a small, highly technical team working on foundational problems in multimodal AI, focused on training models that can interpret, reason, and act on large-scale first-person video data.

You’ll work directly with the Chief Science Officer, shaping how models are designed, trained, and evaluated. The work sits at the intersection of VLMs, long-context reasoning, and real-world deployment.

The focus is on building systems that move beyond static perception, towards temporal understanding, activity recognition, and higher-level reasoning across dynamic environments.

Your work will centre on:

• Designing and training VLMs on large-scale video datasets
• Developing post-training approaches including SFT, RLHF, and parameter-efficient tuning
• Building scalable training and evaluation pipelines
• Exploring long-context and temporal modelling
• Designing efficient systems across edge and server-side inference
• Defining benchmarks for spatial and behavioural understanding

You’ll bring strong experience training deep learning models, ideally transformer-based, along with hands-on work in vision, language, or multimodal systems.

Experience with large datasets, model optimisation, or deploying models into production environments will be valuable. Exposure to video data or long-context modelling is particularly relevant.

This is a team that values speed, ownership, and first-principles thinking. You’ll be working on open-ended problems with real-world impact, with the freedom to explore and define approaches.

Compensation: Highly competitive salary + equity
Location: San Francisco, onsite

If you’re interested in building multimodal systems that operate in real-world settings, and want to join a well-funded, highly skilled research team, please apply now!

All applicants will receive a response.

Ready to own the data pipeline powering the voice of the next generation of AI characters?

You'll be joining a well-funded startup building AI character technology, where speech is a core part of the product experience.

Think super natural conversations, handling interruptions, personality shifts and more!

You'll own the datasets that power their speech systems — from raw, messy audio through to clean, versioned training corpora that directly drive TTS and ASR model performance.

Your focus

• Own the full data lifecycle — defining specs, auditing and curating large-scale audio and text corpora
• Build automated quality metrics and dashboards across SNR, VAD, WER, speaker verification and safety, validated against listening tests
• Train and deploy lightweight classifiers for noise detection, diarisation, language ID, and content moderation

What you'll bring

• Deep experience working with speech and audio data at scale — 1M+ hours
• Strong ML engineering skills in Python and PyTorch, including training and fine-tuning models like Whisper or Wav2Vec
• Practical knowledge of audio processing — torchaudio, librosa, spectrograms, DSP basics
• A solid understanding of audio quality metrics — MOS, WER, PESQ/STOI, SNR, speaker verification

Nice to have

• Experience with Spark/Beam, Airflow, SQL or similar data engineering tools
• Open-source contributions or publications in speech or audio ML
• Background in denoising and enhancement, and how it affects downstream model quality

Remote, with a preference for European or overlapping timezones. Competitive compensation and equity.

Want to build the systems that make AI agents actually work in production?

Most agents fail outside controlled environments, not because the models are weak, but because the systems around them can’t represent how real work happens.

This team is building that missing layer...

Their platform sits inside enterprise workflows, capturing how tasks are executed across tools, then structuring that data so models and agents can actually use it. Real operational context, not synthetic benchmarks.

As a Full Stack Engineer, you’ll focus on the backend and product systems that make this usable in production.

You’ll design workflow data models, build high-throughput pipelines, and ship full-stack features used by real customers. This sits across distributed systems, data engineering, and LLM integrations.

Tech stack includes TypeScript (NestJS, React, Vite, TanStack), PostgreSQL, and AWS/GCP, with OpenAI and Anthropic models integrated into core systems.

You’ll join a small, highly technical, Accel-backed team that’s already post-revenue and scaling with enterprise customers. This isn’t speculative infrastructure, it’s being used.

Experience with Python pipelines, Terraform monorepos, or Rust/Swift is useful, but not essential.

What matters is your ability to build systems that hold up in real-world complexity.

📍 San Francisco (on-site)
 💰 $160K–$280K base + equity + additional comp

If you’re interested in building the layer that makes AI agents usable, this is where that work is happening.

All applicants will receive a response.

Advance how agents and LLMs learn from feedback in realistic environments.

If you've been working at the intersection of reinforcement learning and large language models, this is an opportunity to work on the foundations of how AI systems are trained, evaluated, and supervised, with your research shipping into production.

You will work hands-on on fundamental problems spanning LLM post-training, RL simulation environments, and agentic evaluation, shaping core methods and benchmarks used by leading AI labs and enterprises around the world.

The team actively publishes and collaborates with external research labs, with recent work appearing at ACL and NeurIPS. You'll see your ideas move from concept to deployed systems, working alongside engineers who build fast and take research seriously.

This is a research-driven company growing quickly due to real demand for what they're building. If you want your work to matter, both in the literature and in production, this is where to do it.

You'll bring hands-on experience in applied research across RL, LLM post-training, or agent-based systems, with a strong understanding of transformer architectures and fine-tuning. As important as the theory is the ability to ship — you can translate research ideas into production-ready systems that actually work. A track record of publishing at top-tier venues such as NeurIPS, ICML, ACL, or EMNLP is a plus, but what matters most is the quality of your thinking and your ability to execute.

What you'll do

• Conduct research on LLM post-training methods (RLHF, RLAIF, RLVR)
• Design and build realistic RL simulation environments for agents
• Develop agentic evaluation and supervision frameworks
• Create and maintain benchmarks for emerging AI capabilities
• Collaborate with engineers to take research from idea to deployed systems

Location — San Francisco · Salary — Up to $300k base + equity, flexible and negotiable DOE

All applicants will receive a response.

• PhD and/or publications at top conferences across long-horizon reasoning, RL, or similar
• Post-training experience (RLHF, DPO, reward modelling)
• Experience working on open-ended research 

Want to own the inference layer behind millions of real-world voice AI interactions every day?

You’ll join a profitable, founder-led enterprise conversational AI company powering billions of interactions annually across 30+ languages. Their systems sit behind major global brands and handle millions of customer conversations daily.

They’re now moving toward end-to-end multimodal and speech-to-speech architectures. You’ll own the inference stack powering both their multimodal speech-text LLM and their text reasoning LLM.

This goes well beyond tuning configs.

You will:

• Optimise production inference across A10, A100 and H100 GPUs
• Own scheduler design, KV cache allocation and batching logic
• Build serving systems tailored to multimodal audio-text workloads
• Support agentic, multi-step reasoning under real latency constraints
• Profile kernel-level bottlenecks and fix them properly

You’ve modified inference framework internals before, not just used them. You’re comfortable in Python and C++, and you’re happy diving into CUDA graphs, memory bandwidth limits or custom kernels when required.

This platform processes over 2 million interactions per day. Latency, throughput and cost are production realities, not lab metrics.

Package: €150,000 base + bonus + stock options
Location: Remote within Europe

If you want full ownership of inference performance at real enterprise scale, let’s talk.

All applicants will receive a response.

New York City
$200,000–$250,000 + equity

Define how AI agents actually learn in production.

This team is building the foundational learning framework behind enterprise AI systems.

Not prompt wrappers.
Not repeated fine-tuning.

A system that formalises how work gets done and allows agents to improve continuously in real environments.

You’ll design architectures that turn operational behaviour into structured, executable intelligence — making knowledge compound over time through reasoning loops, persistent memory, and human-in-the-loop feedback, without degrading performance.

You’ll work directly with experienced founders and live enterprise customers on problems where reasoning, context, and workflow execution intersect.

What you’ll work on

• Expanding the core learning framework that governs how agents improve

• Designing structured context and memory layers

• Building reasoning loops and feedback systems

• Creating continuous learning pipelines from live operational data

• Shipping production-grade Python systems into real deployments

What you’ll bring

• Experience building non-trivial LLM systems in production

• Designed agentic workflows involving reasoning, memory, and tool use

• Strong Python engineering and systems thinking

• Clear ownership of end-to-end AI systems

The company

• Series A backed by Sequoia ($28M)

• Platform approaching one trillion tokens processed

• Major enterprise customers live

• Small, engineering-led team building the learning layer enterprise AI will depend on

Everyone will receive a response.

Training builds capability. Post-training decides what it becomes.

This team are rethinking how large multimodal models learn after pre-training — developing post-training and reinforcement learning methods that help models reason, plan, and interact in real time.

Founded by the researchers behind several of the most influential modern AI architectures, this lab are pushing alignment and learning efficiency beyond standard RLHF. They’re scaling preference-based training (RLHF, DPO, hybrid feedback loops) to new model types and creating systems that learn from interaction rather than static data.

You’ll work at the intersection of post-training, RL, and model architecture — designing reward models, scalable evaluation frameworks, and training strategies that make large-scale learning measurable and reliable. It’s applied research with direct impact, supported by serious compute and a tight researcher-to-GPU ratio.

You’ll bring experience in large-scale post-training or reinforcement learning (RLHF, DPO, or SFT pipelines), a solid grasp of LLM or multimodal training systems, and the curiosity to explore new optimisation and alignment methods. A publication record at top venues (NeurIPS, ICLR, ICML, CVPR, ACL) is a plus, but impact matters more than titles.

The team are based in San Francisco, working mostly in person. $1 million+ total compensation. Base salary circa $300K – $600K (negotiable) plus stock and bonus — exact package depends on experience.

If you want to work where post-training meets architecture — shaping how foundation models learn, reason, and adapt — this is that opportunity.

All applicants will receive a response.