AI Engineer II

Mission

We are building the next generation of hyper-personal AI systems that fundamentally transform how humans interact with businesses. Led by our Chief AI Scientist, the Personal Superintelligence Lab at Gopuff develops cutting-edge agentic AI solutions that reason about complex user contexts, preferences, and real-world constraints to deliver superhuman performance in personalized shopping assistance. You will advance the state-of-the-art in prediction, alignment, grounding, and multi-agent orchestration while deploying these breakthroughs at massive scale.

Scope and Impact

You will own core components of our instant shopping personal intelligence—context engineering, grounding, alignment, and serving—so technology accelerates customers to their goals, leveraging our unparalleled delivery speed. You’ll design context engineering strategies and real‑time retrieval, build SFT pipelines, and implement reinforcement learning from human feedback (RLHF) and verifiable rewards (RLVR) using GRPO/GSPO. You’ll use relevant APIs for data augmentation and ship low‑latency, reliable inference (batching, caching, quantization, streaming) with strong observability and safety. In close partnership with Engineering and Data Science, you’ll run experiments end‑to‑end and deliver production models that drive measurable lifts in business KPI under privacy‑by‑design and rigorous evaluation.

Research Areas & Contributions

Advanced Context & Grounding Research:

- Pioneer novel architectures for multi-modal context integration across temporal, spatial, and behavioral dimensions

- Develop foundational techniques for real-time knowledge grounding with dynamic constraint satisfaction

- Research declarative programming paradigms (DSPy) for robust prompt compilation and systematic LLM behavior specification

- Design and develop multi-agent orchestration systems (LangChain, LangGraph, CrewAI, Autogen, LlamaIndex) that exhibit emergent reasoning capabilities

Alignment & Learning Systems:

- Advance reasoning-centered supervised fine-tuning methodologies with novel data curation, synthetic generation, and quality assurance frameworks

- Develop and evaluate model reasoning (task decomposition, tool-use, and self-correction) to improve grounded recommendations and task completion under real-world constraints.

- Research parameter-efficient adaptation techniques including LoRA variants, Text-to-LoRA synthesis, and dynamic adapter routing

- Pioneer RLHF and RLVR with emphasis on scalable oversight

- Develop next-generation policy optimization algorithms beyond GRPO/GSPO with formal safety guarantees

Safety & Robustness:

- Research interpretability, controllability, and alignment verification for production agentic systems

- Develop formal methods for safe exploration, reward hacking prevention, and distributional robustness

- Pioneer privacy-preserving techniques and federated learning approaches for personal AI systems

Systems & Deployment:

- Research efficient inference architectures including novel quantization, caching, and streaming paradigms

- Develop evaluation frameworks that bridge offline metrics with online performance and safety criteria

- Build foundational infrastructure for rapid experimentation and deployment of research breakthroughs

The only predictable thing about life is that it’s wildly unpredictable.That’s where we come in.

When life does what it does best, customers turn to Gopuff to deliver their everyday essentials, and to get through their day & night, work day and weekend.

We’re assembling a team of thinkers, dreamers & risk-takers...the kind of people who know the value of peace of mind in an unpredictable world. (And people who love snacks.)

Like what you’re hearing? Welcome to Gopuff.

The Gopuff Fam is committed to an inclusive workplace where we do not discriminate on the basis of race, sex, gender, national origin, religion, sexual orientation, gender identity, marital or familial status, age, ancestry, disability, genetic information, or any other characteristic protected by applicable laws. We believe in diversity and encourage any qualified individual to apply. We are an equal employment opportunity employer.

#LI-GOPUFF