Coinbase
About this role
Coinbase is the largest listed cryptocurrency exchange in the United States. This role is on the CX Intelligence team, which builds the customer support experiences connecting the Help Center, chatbots and agent workflows, and owns the multi agent platform behind Coinbase chat. Coinbase is explicit that it is hiring at IC5 level. The work is designing and deploying the orchestration layer that coordinates vendor AI, internal multi agent systems and human agents, which means managing state transitions, context sharing and intent routing across a distributed conversational environment. You build production Python services that bridge ML research and reliable customer facing products, lead complex ML projects end to end, set engineering standards for AI and ML workflows, and mentor other engineers. It is fully remote within India, though Coinbase describes itself as remote first rather than remote only.
Who this is for
What the posting asks for: 5 or more years of professional experience in machine learning and software engineering. Coinbase states the level directly, hiring an IC5 Machine Learning Engineer.
The real day to day: architecting and deploying the orchestration layer that manages state transitions, context sharing and intent routing across vendor and internal LLM frameworks in a distributed conversational environment. Building production grade Python services that connect ML and AI research to reliable, measurable customer facing products. Leading end to end execution on complex ML initiatives, managing priorities, technical trade offs and cross functional dependencies from design through delivery. Establishing best practice for system design, coding standards and AI and ML development workflows across the team. Mentoring engineers on architectural integrity and modern AI and ML patterns. Running design reviews.
The team context: CX Intelligence sits inside Coinbase's Enterprise Applications and Architecture organisation and owns the multi agent platform powering Coinbase chat, the Help Center and agent tooling, working with Conversation Design, CX and other engineering teams to deliver secure and compliant AI powered support.
Location and setup, and this matters: the role is remote within India, but Coinbase states plainly that it is remote first, not remote only, and that you should expect to gather quarterly for in person working sessions it calls surges. If travel every quarter is a problem, factor that in now rather than at offer stage. Coinbase is also unusually blunt about culture in the posting itself, saying the bar is high and the environment is intense.
Honest fit guidance: the interesting engineering here is the hand off logic between different LLM frameworks and between AI and human agents, which is a genuinely hard and current problem. It is applied ML with a heavy production systems and architecture component, so it suits someone who has shipped and operated ML services rather than someone whose experience is model training alone. Coinbase has three remote India roles on today's board across different teams.
The real day to day: architecting and deploying the orchestration layer that manages state transitions, context sharing and intent routing across vendor and internal LLM frameworks in a distributed conversational environment. Building production grade Python services that connect ML and AI research to reliable, measurable customer facing products. Leading end to end execution on complex ML initiatives, managing priorities, technical trade offs and cross functional dependencies from design through delivery. Establishing best practice for system design, coding standards and AI and ML development workflows across the team. Mentoring engineers on architectural integrity and modern AI and ML patterns. Running design reviews.
The team context: CX Intelligence sits inside Coinbase's Enterprise Applications and Architecture organisation and owns the multi agent platform powering Coinbase chat, the Help Center and agent tooling, working with Conversation Design, CX and other engineering teams to deliver secure and compliant AI powered support.
Location and setup, and this matters: the role is remote within India, but Coinbase states plainly that it is remote first, not remote only, and that you should expect to gather quarterly for in person working sessions it calls surges. If travel every quarter is a problem, factor that in now rather than at offer stage. Coinbase is also unusually blunt about culture in the posting itself, saying the bar is high and the environment is intense.
Honest fit guidance: the interesting engineering here is the hand off logic between different LLM frameworks and between AI and human agents, which is a genuinely hard and current problem. It is applied ML with a heavy production systems and architecture component, so it suits someone who has shipped and operated ML services rather than someone whose experience is model training alone. Coinbase has three remote India roles on today's board across different teams.
Apply on company site
Opens www.coinbase.com, the employer's own application page. Applying is always free.