עדיין מחפשים עבודה במנועי חיפוש? הגיע הזמן להשתדרג!
במקום לעבור לבד על אלפי מודעות, Jobify מנתחת את קורות החיים שלך ומציגה לך רק משרות שבאמת מתאימות לך.
מעל 85,000 משרות • אלפי משרות חדשות בכל יום
חינם. בלי פרסומות. בלי אותיות קטנות.
DOCX, DOC, PDF - בעברית או באנגלית
כבר העליתם קו"ח בעבר? כניסה
At Innodata, we partner with leading global organizations across a range of industries, providing data, technology, and AI solutions. Among our areas of focus is building platforms and infrastructure that enable leading AI labs to develop, train, and evaluate the next generation of AI models and autonomous agents.
We are looking for a Senior Full-Stack Platform Engineer to take a leading technical role in building scalable platforms, executable environments, and evaluation systems that help researchers understand, test, and measure how AI agents perform in complex, real-world scenarios.
This is an opportunity to work at the intersection of software engineering, AI research, and agentic systems, tackling complex technical challenges in a rapidly evolving field.
Location: Fully Remote – Israel
About the role
You will be the senior engineer on the platforms that produce executable data products for frontier AI labs: environments agents act in, tasks that run, verifiers that score, and the pipelines and interfaces that let humans and models judge the result.
This is a platform role, not a feature role. What you build is used by client researchers, by the raters and domain experts producing judgments, and by our own delivery team. Much of the work is client-facing, supporting the client's product and research teams directly.
What you'll do
- Build the platform itself: shared services, data models and interfaces that let new task types, environments and grading approaches be added without rebuilding the system.
- Build executable data products: simulated environments with controllable state, task generators, verifiers and graders, trace capture and replay - data that runs, is versioned, and re-runs deterministically months later.
- Own the full stack: backend services and pipelines, the data layer underneath, and the UI on top.
- Integrate models on both the creation and execution side: model-backed task generation, model-as-judge grading calibrated against humans, and agent harnesses that run for hours. You design for components that do not behave the same way twice, yet must be evaluated fairly every time.
- Take prototypes to production: get a new environment type or grading approach working in days, test it with real tasks and users, and harden it into a platform capability when it proves out.
- Write design docs and operating notes that researchers and non-engineers can follow.
What we're looking for
- 10+ years building and shipping production software, with substantial full-stack experience: backend services, data layer and user-facing interfaces.
- Built or owned a platform: shared infrastructure, internal tooling, an SDK, or a product other teams built on top of.
- Shipped executable data products - pipelines, environments, harnesses, simulations or evaluation systems - where the deliverable was code and data together.
- Hands-on work with language models and agents in production: tool calling, multi-step traces, prompt and harness iteration, cost and reliability management.
- Comfortable using AI coding agents in your daily development workflow, with opinions on where they help and where they mislead.
- Product-oriented: you ask what a researcher is trying to learn and what decision the output will drive, and design backward from that.
- Clear written communication for an audience of researchers, raters and engineers.
- Interest in taking on problems nobody has solved yet: you can commit to an approach, learn it was wrong, and change course without losing momentum.
Nice to have
- Agentic evals experience: environments, verifiers, trajectory grading, model-as-judge calibration or benchmark construction. Candidates with this experience will be prioritized.
- Model training or research experience; you will work closely with ML and RL researchers.
- Deterministic replay, event sourcing or simulation state management.
- Workflow orchestration for long-running jobs (Temporal or equivalent).
- Sandboxing and multi-tenant isolation.
- Data-dense frontend work: rendering long structured traces legibly.
- Privacy engineering: PII handling, de-identification, audit trails.
About us
With more than 2,000 customers and operations in 13 cities around the world, Innodata is the AI technology solutions provider-of-choice for 7 of the world’s biggest technology companies, as well as leading companies across financial services, insurance, law, and medicine. We’re at an exciting stage in the company and fortunate in these times to grow our team.
We’re an equal opportunity employer.
Only relevant applicants will be contacted
במקום לעבור לבד על אלפי מודעות, Jobify מנתחת את קורות החיים שלך ומציגה לך רק משרות שבאמת מתאימות לך.
מעל 85,000 משרות • אלפי משרות חדשות בכל יום
חינם. בלי פרסומות. בלי אותיות קטנות.
DOCX, DOC, PDF - בעברית או באנגלית
כבר העליתם קו"ח בעבר? כניסה
שאלות ותשובות עבור משרת Senior Full-Stack Developer, Agent Evaluation Platform
התפקיד כולל פיתוח של פלטפורמות מורכבות להערכת סוכני AI, בניית סביבות הרצה בנות-קיימא, מערכות דירוג ואימות, וניהול מלא של ה-Full-Stack החל מהצד האחורי (Backend) ועד לממשק המשתמש, במטרה לתמוך בצוותי מחקר מובילים בתחום ה-AI.
משרות נוספות מומלצות עבורך
-
*Senior Full-Stack Engineer
-
תל אביב - יפו
Buildots
-
-
Senior Full-Stack Engineer
-
ישראל
TITAN
-
-
Senior Full-Stack Developer
-
תל אביב - יפו
Madlan
-
-
Senior Full-Stack Engineer
-
רמת גן
Active Fence
-
-
Senior Full-Stack Engineer – Mobile & Backend
-
תל אביב - יפו
New Como Ltd
-
-
Senior Full-Stack Engineer
-
רחובות
Medulla
-