AI Studio Coach- Agentic AI

Break Through Tech•New York, NY

2d•Remote

About The Position

Break Through Tech is seeking subject matter experts to support delivery of its Agentic AI programs for up to 500 undergraduate fellows. Coaches are assigned to the Beginner or Advanced level based on their experience and staffing needs. The 16-week program builds on fellows’ foundations in machine learning and Python and teaches them to design, build, evaluate, and deploy agentic systems. Fellows work in teams, using LangGraph to build agents, LangSmith to trace and evaluate them, and cloud platforms to deploy their systems. The program includes four units covering agent fundamentals, architecture, evaluation, and deployment, with asynchronous coursework completed before monthly virtual Workshops and Maker Days. Each Coach supports 8–10 project teams, approximately 40 fellows. Coaches facilitate virtual events, provide technical mentorship and code review, conduct recurring team check-ins, monitor progress, assess deliverables, and give feedback. Advanced Coaches additionally mentor fellows on multi-agent architectures and coordination patterns. This is a part-time, remote, temporary position running from January 4, 2027, through May 20, 2027. The estimated commitment is 15–20 hours per week, with weekly hours varying according to program activities. Coaches must complete about 10 hours of pre-program training in early January and be available for eight program events from January through May on dates to be determined. The role pays $35 per hour and is not eligible for benefits. It is open to eligible candidates across the United States; visa sponsorship is unavailable, though candidates authorized to work through F-1 CPT or OPT may apply.

Requirements

Bachelor’s or Associate’s degree in computer science, data science, or a related field.
Strong background in AI/ML with hands-on experience in agent development.
Personal experience building an agent with tools and/or RAG and evaluating its performance.
Hands-on experience with LangChain and LangGraph.
Hands-on experience with RAG systems, API integrations, and deploying applications to cloud platforms.
Experience with evaluation and observability/tracing tools, particularly LangSmith.
Experience with deployment and containerization, including Docker and cloud platforms.
Prior teaching, tutoring, or mentoring experience, including coaching or leading a team through an ambiguous, open-ended technical project.
Experience giving feedback on presentations or providing public-speaking coaching.
Solid knowledge of agent architecture, including tool calling, RAG, and memory.
Comfort reading, writing, and debugging Python code; ability to troubleshoot code and debug complex systems.
Excellent communication and facilitation skills, including evaluating and giving feedback on technical work.
Ability to facilitate a live session for a group of about 40 fellows.
Ability to thrive in a virtual work environment.
Commitment to developing inclusive and equitable working relationships with fellows from a variety of backgrounds, volunteers, other Coaches, and program staff.
Availability for approximately 15–20 hours per week, about 10 hours of pre-program training in early January 2027, and eight program events from January through May 2027.

Nice To Haves

An advanced degree.
For Advanced Coaches: solid knowledge of multi-agent coordination patterns, including supervisor/orchestrator workflows, agent handoffs, and agents-as-tools.
For Advanced Coaches: experience building a multi-agent system or a single-agent system with advanced tool routing, and running systematic evaluations of agent performance.
For Advanced Coaches: hands-on experience with LangGraph subgraphs, parallel execution, and multi-agent coordination patterns.
Prior Break Through Tech and/or Machine Learning Foundations experience.

Responsibilities

Facilitate four monthly 1.5-hour Workshops from start to finish, including code demonstrations and guided project work sessions.
Facilitate four monthly Maker Days from start to finish, structuring build sprints, checking milestone progress, and troubleshooting technical issues live.
Prepare for each Workshop and Maker Day.
Maintain attendance records for Workshops, Maker Days, and team check-ins.
Mentor teams on agent design, RAG and tool integration, evaluation and observability with LangSmith, and cloud deployment and containerization.
For Advanced Coaches, mentor teams on multi-agent architecture and coordination patterns.
Review code and conduct debugging sessions to troubleshoot complex technical issues.
Find and recommend additional resources to support team projects as needed.
Attend and actively participate in twice-monthly Community of Practice meetings with Coaches and program staff.
Hold a recurring check-in at least monthly with each of 8–10 assigned teams to monitor progress and provide feedback.
Monitor team progress and intervene to address technical blockers or team-dynamics issues.
Partner with program staff to manage unresolved issues affecting project success.
Conduct outreach to at-risk fellows, including possible one-to-one meetings and post-meeting follow-ups.
Grade unit and event deliverables for assigned teams using provided rubrics.
Track deliverable submissions and follow up with teams that have not submitted.
Provide constructive technical feedback on project milestones and presentations.
Observe final presentations and provide meaningful feedback.