This is a remote role.

This is a 3 months contract role.

Role Summary

Our client is building SuperHive, an AI-powered enterprise platform that uses agentic AI to transform complex domain data into structured, actionable intelligence. The platform spans multiple engineering squads, a new infrastructure layer (workflow engines, agentic orchestration, knowledge pipelines), and integrations with third-party ecosystems. We need someone to make sure all of it ships.

As a Senior Technical Program Manager, you will own the execution of SuperHive — a platform rebuild, new AI infrastructure, and external partner engagements. You will be embedded in the build, not observing it from a dashboard. You will sit in architecture discussions, challenge timelines with informed opinions, manage the critical path across squads, and keep a fast-moving team aligned without slowing it down.

The TPM function at our client is lean and goes deep where it matters. You carry two things at once: teams you are embedded in, and cross-team initiatives you own end-to-end. The first keeps you grounded in the build. The second is what makes the function strategic. You need to be technically fluent enough to understand why an agent pipeline is taking longer than planned, to spot a dependency risk in an API contract discussion, and to know the difference between it works in a notebook and it works in production.

Key Responsibilities

Execute

Own the end-to-end delivery plan for SuperHive, coordinating dependencies, milestones, and integration points.
Break down the platform roadmap into quarterly delivery plans with clear squad-level objectives, measurable outcomes, and realistic timelines — accounting for the inherent uncertainty in AI/ML feature development.
Partner with Tech Leads and Product Managers daily: enhance clarity on scope, surface dependencies and risks early from PRDs and technical designs, and keep all the parts together.
Run the program rhythm: sprint cadences, cross-squad syncs, demo cycles, retrospectives. Keep them lean and useful — no ceremony for ceremony's sake.
Manage the critical path. When one squad is blocked waiting on another, you find the path forward — whether that means re-prioritizing, descoping, or pulling in help.

Influence

Own and drive at least one complex cross-functional program end-to-end: the kind of initiative that touches multiple squads, spans Engineering and Product and leadership, and would not happen without someone holding it together.
Facilitate technical alignment when squads need to agree on API contracts, data schemas, event formats, or shared infrastructure. You do not decide the architecture — you make sure the right people are in the room and the decision gets made and documented.
Manage external prototyping and development engagements: charter and scope definition, milestone tracking, integration checkpoints, and acceptance criteria. External partners build features that must integrate cleanly with the platform, and you own that lifecycle.
Challenge the status quo. If a process is not working, propose a better one. If a timeline is unrealistic, say so with data.

Cultivate

Contribute to building the TPM function itself: shape ways-of-working rituals, bring new ideas for how the team operates, and represent what the function does across the organization.
Build visibility into delivery health for engineering leadership — dashboards, weekly updates, or whatever format actually gets read and acted on. Honest, data-informed, no sugar-coating.
Define and enforce release processes: what needs to be true before a feature ships (tests, security, documentation, observability), without making the process a bottleneck.
Contribute to our clien's hiring and branding initiatives for the Engineering function.

AI/ML Delivery Awareness

A significant part of what makes this role different from a generic TPM position is the AI/ML delivery context. You will need to:

Understand the non-linear nature of AI workstreams: evaluation-driven iteration, prompt engineering cycles, model performance plateaus, accuracy improvements that come through data and eval loops rather than just feature completion.
Work with the Intelligence squad to plan agentic AI delivery in phases — knowing that a multi-agent system is not done when the code is written, but when the evaluation benchmarks pass.
Track AI-specific metrics alongside delivery metrics: model accuracy, latency budgets, token costs, evaluation scores. Recognize when these metrics signal a timeline risk.
Navigate the tension between move fast and get the AI right — helping squads make pragmatic trade-offs between shipping and iterating on quality.

Required Qualifications

7+ years of experience in Technical Program Management, Engineering Management, or an adjacent role where you coordinated technical delivery across multiple teams in a fast-paced environment.
At least 2–3 years managing programs that included AI/ML workstreams shipping to production — not research or proof-of-concepts, but products with real users.
Demonstrated experience coordinating 4+ engineering teams working on a shared platform with real inter-service dependencies.
Technical fluency: you can read an architecture diagram, understand an API contract discussion, ask informed questions about database schema choices, and recognize when a timeline estimate is optimistic versus grounded. You do not write code, but you speak the language.
Startup pace: you have operated in environments where scope changes frequently, teams are small, and you balance speed with quality by making smart trade-offs — not by adding process.
Strong understanding of modern software delivery: CI/CD, microservices, containerization, cloud infrastructure (AWS preferred).
Excellent written and verbal communication. You write clear status updates that engineering leadership actually reads. You run efficient meetings. You translate technical complexity into executive-friendly language without losing accuracy.
Experience managing external vendor or partner engagements with clear deliverables, integration checkpoints, and acceptance criteria.

Preferred Experience

Direct experience with LLM-powered product delivery: understanding evaluation frameworks, RAG pipeline timelines, prompt iteration cycles, and the operational challenges of running LLM agents in production.
Familiarity with agentic AI architectures: multi-agent systems, tool-calling patterns, orchestration frameworks (LangGraph, LangChain), and what makes them hard to ship reliably.
Experience with workflow orchestration tools (Hatchet, Temporal, Airflow) and message brokers (RabbitMQ, Kafka) — understanding how data pipelines are built and where they stall.
Background in B2B or enterprise platform products with multi-tenancy, role-based access, billing, and compliance requirements.
Experience setting up the TPM or program management function from scratch — defining the cadences, tools, and reporting that actually work rather than inheriting an existing PMO.
Familiarity with observability tools (Sentry, Datadog, Signoz) — understanding how production health informs delivery decisions.

Who You Are

Beyond experience, here are the qualities that matter most in this role:

Resourcefulness: you thrive when dropped into a new context and figuring out your own way forward. You are comfortable where the playbook is still being written, and you shape the function rather than wait to be handed one.
Technical curiosity: you can dive into AI concepts and hold substantive discussions with engineers — critiquing designs, surfacing risks, asking the question that makes a tech lead pause.
Ownership: you can point to complex programs that would not have shipped without you. Not facilitating — owning.
Soft power: you influence without authority, push back without making it personal, and stay credible with engineers, PMs, and leadership at the same time.
Clarity: you connect the dots others miss, especially around cross-team dependencies and organizational bottlenecks. You reach alignment through shared understanding, clear writing, and consistent direction.