[Project Hive]
About the company: 

We’re building the one place for accurate information on the web. In our decentralized platform, accurate information always wins against other noises. Current approaches to determine accuracy don’t work ‒ AI is still ineffective in reasoning, and naive human moderation is biased and not scalable. We are building a novel approach that combines the power of ML/NLP (to rigorously deconstruct discourses) and probabilistic graphical models (PGM) (to drive collective behaviors towards reason at scale), with intuitive and highly usable UX.

We are currently still in stealth mode, with a product in alpha stage and sufficient funding, backed by experienced entrepreneurs, media veterans, and a deeply technical team in software engineering, machine learning, and mathematical sciences. The current team of full-timers and part-timers hails from NTU, NUS, Stanford, Harvard, and other top institutions. The team is currently distributed across three continents, with the tech team mostly based in Singapore.

Job Description and Requirements:  

You will be a key contributor to key problems that make our product work towards harnessing crowd behavior towards producing accurate information, namely discourse parsing and intent detection (based on a rigorous discourse ontology), discourse semantic embedding, and various NLP tasks related to query search and claim-similarity search (including authoring rules based on constituency/dependency parsing, in combination with machine learning).

As an early team member, you play an important role in shaping the scientific rigor in the company and establishing work best practices and tooling.

You might also participate in product planning, user testing, and other activities to help you understand the market context and implementation of your work.


- Good knowledge of computational linguistic or traditional linguistics
- Good understanding of statistics and fundamentals of machine learning
- Strong analytical skills and critical thinking for modeling and validating models in various problem domains
- Familiarity with Jupyter notebook environments and machine learning and NLP libraries
- Coding skills, preferably in Python, Java, or Go
- Ability to work in a more ambiguous situation and communicate with non-academic co-workers and end-users

We accept full-time or part-time candidates, interns, and remote candidates, as long as you have the right skill set.

We will support you with the necessary purchase of data, data labeling service, compute infrastructure, and other resources necessary for you to do the work well.

We allow you to write publications related to your work given some boundary for protecting our intellectual property.

Job Type

Full Time





Closing Date

Apr 30 2020