Morph

Goal: 99.99% uptime We serve custom inference stacks that have irregular GPU load. We're looking for people that have done genuinely amazing work in infrastructure that are interested in a challenge, working with both traditional infrastructure such as load balancers, NLB, etc., as well as very different infrastructure around inference engines and GPU loads. This is a role that will inherently require deep experience with inference engines. Contributions to vLLM, SGLang, trtllm, or inference frameworks a plus. Every role at Morph comes with unlimited tokens on claude code/codex

Company: Morph Location: San Francisco, in person Comp: Base + commission + meaningful equity About Morph Morph builds fast models for coding agents. Our models help agents search code, apply edits, compress context, and run faster in production. We sell to AI coding companies, devtools companies, and engineering teams building with agents. Morph’s hot takes: Being elite at context switching matters - 1 sufficiently person can do the job of 10 people now if they’re AI pilled enough We’ll be the first sub 10 person $1b company All agents will become coding agents The role This is a hands-on sales role and you’ll will work directly with the founder to turn founder-led sales into a repeatable GTM motion. Your job is to create pipeline, run discovery, demo the product, close customers, and write down what works. What you’ll do Own outbound and inbound sales Sell to technical founders, AI teams, devtools companies, and engineering leaders, run discovery, demos, follow-ups, and close plans Build target account lists Turn customer conversations into better messaging Help define ICP, pricing, objections, qualification, and sales process Feed product learnings back to engineering Build the first Morph GTM playbook Who you are You are a technical founding seller. You may be: A strong AE at a technical B2B startup A sales engineer who wants to own revenue A former founder who has sold technical products A technical person with unusually good sales instincts An early GTM hire who wants to build the first motion, not inherit one You should be comfortable talking about APIs, developer workflows, AI coding agents, latency, reliability, deployment, and enterprise engineering constraints. What we care about You can build pipeline, close customers, and sell to technical buyers You follow up hard, write clearly, and learn fast You have high agency, thrive without structure, and want ownership You are willing to do unscalable things What we do not want You need a system set up before you can sell You are strongest at scaling someone else’s machine You prefer strategy over customer calls Why Morph instead of somewhere else: build state of the art GTM software/slack apps (founder already has) Top 0.0001% engineers will teach you how to code Meaningful equity First 90 days In your first 30 days, you will: Learn the product, sit in on founder-led sales calls, and talk to customers Build account lists, start outbound, and take over follow-up Write the first version of our ICP and messaging Own a segment of pipeline, run discovery and demos, and create qualified opportunities Close founder-assisted deals and improve the sales process In your first 90 days, you will: Close revenue without the founder driving every step Show a repeatable path to pipeline Know which customers to focus on and which objections matter Help decide the next GTM hire Interview process We will ask you to: Walk through your past sales numbers Review a real customer call Pitch Morph after short prep Write a simple 30/60/90 plan We care about slope, hunger, judgment, and whether you can sell to companies with coding agents Apply Send whatever best shows that you can sell technical products and learn fast. Past numbers, deal stories, customer emails, outbound examples, founder references, or a short note are all useful.

Morph builds the fastest LLM code-editing inference engine in the world. We hit 10,500 tok/sec per request on NVIDIA hardware. Our stack powers high-throughput AI workflows for vibe coding apps, devtools, PR bots, and IDEs. We’re hiring a founding ML Researcher to push the limits of model capability, throughput, and reliability across inference, retrieval, and edit application. This is a research role that ships. If your work cannot survive contact with production, it does not count here. We’re looking for someone with broad, T-shaped spikey experience across research, systems, and product, plus a deep spike in modern LLM training and inference. You bring taste and judgment. AI can accelerate execution. It cannot replace those. What You’ll Do Design and run experiments for LLMs specialized for code workflows: retrieval, search, editing, and tool use Train and fine-tune models (SFT + preference / RL variants), build evals, and close the loop until results are real Turn new research into production: model packaging, serving constraints, latency budgets, failure modes, monitoring Work directly on inference performance when it matters: KV cache strategy, batching, quantization, speculative decoding, kernel level bottlenecks Collaborate on data strategy: high signal datasets, preference data formats, automatic labeling, and rigorous evaluation You’re a Fit If You PhD level or equivalent experience with PyTorch (plus TF or JAX is fine) Can implement papers without cargo culting them, and can explain why they work. Understands how to distiguish between papers that are noise and real Have shipped ML systems that run under real constraints: latency, cost, reliability, observability Understand modern LLM training mechanics and tradeoffs (data, objectives, RL, evals, inference) Prefer ownership and agency over committees and process theater Bonus Points Experience with CUDA, kernels, Triton, TensorRT-LLM, vLLM, or custom inference stacks Experience with retrieval systems (embeddings, reranking, indexing) and evaluation methodology You have strong opinions about what matters in ML, and can defend them with evidence Why Morph Zero fluff. Work directly with the founder. Everyone on the team is an ML engineer No busywork. If it doesn’t move the needle, we don’t do it Work on the fastest coding subagents in the world, and the research that makes it faster and smarter Apply Describe the ML project you’re most proud of. Go deep on modeling choices, training setup, data, evals, failure cases, and what you’d do differently - the founder reviews every application personally and is a former ML engineer Describe what you’re deeply obsessed with (anything). We care about intensity and taste Every role at Morph comes with unlimited tokens on claude code/codex