Clara Labs is establishing a new class of virtual assistant that understands you like a person, but operates at the scale, speed, and persistence of a machine. To do this, we mix intelligent automation software with remote human contractors to form an efficient, distributed remote-knowledge-work service.
Our initial focus is on scheduling meetings via email. We believe that by reducing anxieties caused by cluttered inboxes, complex calendars, and juggling preferences and calendars, customers can have more time to focus on what matters.
We aim to make automated assistants a truly everyday productivity tool. Our approach differs from traditional machine assistants (think: Siri) in our deep focus on conversation quality and human-like product flexibility. We build technology to maximize user experience, without constraining ourselves to the latest technology bandwagon. As a result, we’ve been delivering a highly reliable and flexible natural language interface from day one.
You can read a little bit more about us here:
Engineering at Clara
Shipping early and often leads to better products.
As an example, we scheduled thousands of meetings for our first customers manually before building any software. We’ve found that persistent iteration leads to tighter feedback loops between our customers, our contractors, and our product development team.
Ownership enables contributors to do their best work.
Critical thinking and invention are key virtues of each team member. Individuals identify problems and solutions, and are responsible for driving projects to completion.
We're taking on problems that are previously unsolved: building large-scale human-machine learning hybrid systems. Here are a few types of problems engineers at Clara work on:
• Semantic interpretation of received messages and calendar events; includes conversation modeling and natural language processing.
• New user interfaces for combining machine learning and human expertise to process message-based tasks; we’ve built a “mail client” and “calendar client” like no other.
Read more about how our machine learning development works here:
You will work on soup-to-nuts development of novel algorithms informed by our unique applications and constraints. Individuals in this role design data collection strategies, frame the right problems to solve, develop models, measure and compare model performance, and integrate these models into production features.
ML team members embed with a product-focused engineering team to ensure that:
• ML predictions are relevant and usable within the primary platform;
• confidence metrics can be integrated into automation systems; and
• data/annotation collection facilities for each problem are baked into our platform.
You have experience in one of the following and familiarity with several others:
• NLP / computational linguistics: techniques (e.g., pos tagging, dependency parsing, chunking, classification, ...) and tools (e.g., Stanford NLP, nltk, …).
• Bayesian inference: e.g., topic modeling, graphical generative models.
• ML methods: e.g., svms, random forests, convex optimization, transfer learning.
• Deep learning: rnn / cnn architectures, algorithms (e.g., rmsprop, adadelta, adam, ...), and tools (e.g., theano, torch, tensorflow, …).
• ML in practice: e.g., selection bias mitigation, ROC analysis, field performance analysis, data mining.
• ML systems: event-driven real-time ML systems, pipelining and processing frameworks (e.g., Spark, Lucene, EMR).
Nice to have(s):
• M.S. or Ph.D. in CS, EE, or equivalent.
• Industry experience with ML systems is strongly preferred.
Communicating machine learning results and capabilities are important on this team. Please include a cover letter explaining why you think you are a great match for Clara.
We are an equal opportunity employer and value diversity at our company. We do not discriminate on the basis of race, religion, color, national origin, gender, sexual orientation, age, marital status, veteran status, or disability status.Read Full Description