AI Engineer

Worldwide (Remote) • Full time

  • Machine Learning

Forward Labs ( is building superhumans, autonomous AI agents for global freight — the backbone of global economy — a trillion-dollar industry. (think for freight)

The company was founded in San Francisco by 3 friends from Xendit (YC S15, 1st YC unicorn in Asia from Berkeley) — early employees of an exited AI startup & 3 Y Combinator-backed startups from 0 to $1b+.

We’re now a fully remote team 🇺🇲🇨🇦🇰🇷🇵🇭🇸🇬. Our team is still quite small (<10) — so you’ll have a massive impact on the trajectory of the business from Day One + working with the founders directly and learning about Silicon Valley & emerging market experience.

We’re backed by world-class VCs from Silicon Valley, Canada, and Korea (including the ones backed by Jeff Bezos and early @ Uber).

What we’re building

Our platform, Forward AI, allows forwarders to automate their processes from booking to tracking and billing - so they can grow their businesses to move goods seamlessly without hiring more staff

Forwarders lose business when they book late. It takes 10-40 emails to book a cargo. Human Agents take 1-5 days on each booking — our AI Agent cuts it to minutes. It cuts costs too, by procuring from 10+ carriers instead of 3. This adds up to 1000s of cargos each day.

Specifically, provision it with a corporate email, and our AI Agent takes care of inquiry, approval, quotation, procurement, and billing. It runs on GPT to talk to shippers, employees & carriers. And, we built a backend OS to pull pricing data & log new bookings like a TMS.


  • Work closely with various stakeholders (primarily customers) to understand and translate their needs into technical specifications
  • Establish a systematic approach to collect and manage extensive datasets of logistics conversations (texts, PDFs, images, tables, etc)
  • Evaluate LLMs/OCR models, whether third-party closed-source or open-source solutions, and recommend viable design solutions
  • Prototype ML models and pipelines to align with business objectives
  • Orchestrate, develop, and deliver end-to-end AI solutions - from architecture design, prototyping to testing, production-grade deployment, documentation, and maintenance
  • Train, prompt/fine-tune, deploy, and evaluate LLMs
  • Stay up-to-date on the latest AI trends, including emerging prompting techniques, Retrieval-Augmented Generation (RAG) strategies, and agentic AI frameworks. But discerning and strategic when evaluating new technology for commercial application
  • Deep-dive into token optimizations, preventing hallucinations and other exceptions

Example projects

  • Human-in-the-loop platform enhacement: to streamline QA processes, improving efficiency and decision-making
  • Logistics task intent classifier: to better understand and respond to customer inquiries.
  • LLM/RAG optimizations: to extract information from unstructured & structured data (PDFs and tables) more accurately and efficiently within a bespoke framework.
  • Large Agentic Model (LAM): to create multi-agentic AI that learns, reasons, and predicts actions based on a comprehensive knowledge base and human feedback
  • MLOps: to establish and streamline a data training and model alignment process, making it accessible for non-technical team members.
  • Model Experimentation: to design a regression framework to evaluate and speed up model experimentation, ensuring rapid innovation
  • AI Guardrails: to ensure ethical and safe use of technology.

Tech stack

We are tech-stack agnostic in terms of the skill sets we are looking for, but the tech stack looks like the below (expect this to change as we grow):

  • LLMs: OpenAI, Hugging Face, Mistral, DocumentAI
  • Backend: Python, FastAPI, Typescript, Supabase
  • Frontend: Next.js, GraphQL, Refine
  • Infrastructure: AWS, GCP, Docker

A great fit, some or all of the followings

  • Strong understanding of ML/AI concepts (NLP, OCR, LLMs, model training, finetuning, GPT, embeddings, etc)
  • Familiar with LLMs through APIs (OpenAI, etc) and with standing up your own models to run inference
  • Is scrappy, comfortable with ambiguity, taking end-to-end ownership to ship, and values progress over perfection
  • Wants to set, maintain, and adhere to great software engineering standards (architecture review, comments, testing, documentation)
  • Never bored of spending time on your favorite newsfeed (e.g. Twitter, GitHub) browsing the latest models and research papers
  • Are enthusiastic about the vertical applications of AI/LLMs in logistics & future of work in particular
  • Comfortable communicating in English

Bonus points

  • Experience working with AI/freight startups
  • Familiarity with AWS & GCP
  • Strong remote working communication skills

Interview Process

We expect the process to be complete within weeks.

  1. Video Call with founder(s): Values, culture-fit, getting to know more about you and answer your questions about Forward Labs.
  2. [If needed] Take-home assignment / sample work submission
  3. Paid work trial (2 days to 1 week depending on availability and scope): to help you get a sense of how it feels to work with Forward Labs and vice versa. This will include, but not limited to - system architecture, prototyping, demo

Email for more details