Sayantan Das
Senior Applied AI Scientist,
Manulife

ABOUT THE SPEAKER:

I’m a Senior Applied AI Scientist at Manulife, where I work as an internal Forward Deployed Engineer — sitting close to the business and closer to the metal — building agentic AI products tied to real business outcomes.

Previously, I owned RLOps from thought to production at Prem Labs for their agent finetuning platform, now live on AWS Marketplace. I’m an early contributor to Hugging Face’s trl library, the most widely used LLM finetuning library in the world.

I completed my Master in Applied Science from Queen’s University under the supervision of Dr. Ali Etemad. My research on Deepfake Detection was partially funded by the Vector Institute Scholarship in AI. Published in NeurIPS and TMLR. Reviewer at NeurIPS and ICASSP 2026. Listen to this podcast to know more.

In the 5pm–9am part of my life, I advise and help traditional businesses become AI-ready.

TALK TITLE:

Loop Engineering is just K8s-style cybernetics

TRACK:

Agent Memory Architectures

SUB TOPIC:

TBA

ABSTRACT:

This session reframes “Loop Engineering” as applied cybernetics — the same observe→diff→act reconciliation pattern that powers Kubernetes controllers, now applied to AI agent loops. Attendees will learn how to map K8s design principles (declarative goals, idempotent actions, crash-only design, admission webhooks as HITL gates) directly onto agent infrastructure, and walk away with a concrete checklist for building production agent loops that self-heal, externalize state, and fail gracefully. We’ll cover real failure modes — thrashing, stale reads, infinite reconciliation, conflicting controllers — and the fixes that transfer from a decade of K8s production learnings.

WHAT YOU’LL LEARN:

  • Define exit conditions, not paths. “All tests pass” is a reconciliation target; a 47-step plan is fragile.
  • Make every action idempotent. If the loop re-executes a step, nothing should break or duplicate.
  • Externalize state aggressively. The agent should be killable and another instance should pick up seamlessly.
  • Observe before every action. Never trust cached context — always re-read current reality.
  • Add backoff and damping. Loops that over-correct on each iteration thrash (same as K8s pod autoscaler oscillation).
  • Set hard ceilings. Max iterations, cost budget, and human gates prevent infinite reconciliation on unreachable goals.
  • Scope ownership clearly. Two loops editing the same file without coordination will conflict — use labels/task-scoping just like K8s selectors.

Who Attends

Attendees
0 +
Data Practitioners
0 %
Researchers/Academics
0 %
Business Leaders
0 %

2023 Event Demographics

Technical practitioners working directly with ML/AI systems
0 %
Currently Working in Industry*
0 %
Attendees Looking for Solutions
0 %
Currently Hiring
0 %
Attendees Actively Job-Searching
0 %

2023 Technical Background

Expert/Researcher
14%
Advanced
37%
Intermediate
28%
Beginner
7%

2023 Attendees & Thought Leadership

Attendees
0 +
Speakers
0 +
Company Sponsors
0 +

Business Leaders: C-Level Executives, Project Managers, and Product Owners will get to explore best practices, methodologies, principles, and practices for achieving ROI.

Engineers, Researchers, Data Practitioners: Will get a better understanding of the challenges, solutions, and ideas being offered via breakouts & workshops on Natural Language Processing, Neural Nets, Reinforcement Learning, Generative Adversarial Networks (GANs), Evolution Strategies, AutoML, and more.

Job Seekers: Will have the opportunity to network virtually and meet over 30+ Top Al Companies.

Ignite what is an Ignite Talk?

Ignite is an innovative and fast-paced style used to deliver a concise presentation.

During an Ignite Talk, presenters discuss their research using 20 image-centric slides which automatically advance every 15 seconds.

The result is a fun and engaging five-minute presentation.

You can see all our speakers and full agenda here

Get our official conference app
For Blackberry or Windows Phone, Click here
For feature details, visit Whova