UPDATE: OpenAI’s head of forward-deployed engineering, Colin Jarvis, just revealed transformative strategies to help major companies transition from AI hype to real-world applications. His team is embedding itself within organizations to unlock significant value from AI, with projects generating “tens of millions to sometimes the low billions” in value.
In a recent episode of the “Altimeter Capital” podcast, published on July 13, 2023, Jarvis explained how his small but growing team of 39 engineers is set to expand to 52 engineers by year-end. OpenAI currently lists 24 job openings for forward-deployed engineers across major cities including San Francisco, New York, Dublin, and London, with U.S. salaries reaching up to $345,000 plus equity.
The forward-deployed engineering model, inspired by Palantir, involves engineers collaborating directly with clients to tailor AI technologies to their specific needs. Jarvis noted that while the launch of ChatGPT in 2022 ignited excitement, many enterprise customers struggled to harness its potential. He emphasized that embedding engineers directly with clients is crucial for success, stating, “the only consistently successful approach was to embed directly with clients.”
One standout project involved Morgan Stanley, a pioneer in deploying GPT-4. The technical setup took six to eight weeks, but gaining the trust of financial advisors required an additional four months of pilot programs and evaluations. Ultimately, Jarvis reported that about 98% of the advisors adopted the new technology, showcasing the model’s effectiveness.
Jarvis also shared insights from a project with a European semiconductor company, where engineers were found to spend 70% to 80% of their time debugging chips. The forward-deployed team created a “debug investigation and triage agent” that streamlined this process, highlighting the significant impact of their interventions.
In his announcement earlier this year, Jarvis outlined the mission of his team: “Our focus is getting our customers to production, whether it’s through a zero-to-one novel application of our tech or helping you to scale proven cases.”
As OpenAI ramps up its hiring efforts globally, Jarvis pointed out that the forward-deployed engineering model is pivotal to accelerating AI technology into production. Oliver Jay, OpenAI’s international managing director, echoed this sentiment at the Fortune Brainstorm AI 2025 conference, stating that it addresses critical gaps between technology and its application in businesses.
This innovative approach is capturing the attention of venture investors as well. Diana Hu, a partner at YC, mentioned that founders have successfully closed “six, seven-figure deals” through forward-deployed engineering. Garry Tan, YC’s CEO, noted that this model gives AI startups a competitive edge over larger firms like Salesforce and Oracle.
As AI continues to evolve, OpenAI’s forward-deployed engineering team is positioned at the forefront of this revolution, driving immediate and impactful change across industries. Stay tuned for more updates on how this model reshapes the future of AI adoption.