Forward Deployed Engineer (FDE)

TL;DR

Role  Forward Deployed Engineer (FDE)

Also known as  Solutions Engineer · Technical Field Engineer · Implementation Engineer

Location  Paris — hybrid (regular on-site presence at client locations)

Contract  Permanent contract within an OSS Ventures portfolio startup

Compensation  To be defined based on startup and profile

Seniority  5+ years of hands-on experience required

Start date  Ongoing — based on portfolio openings

About OSS Ventures

OSS Ventures is an industrial startup studio based in Paris. We build and co-develop B2B startups that use AI and data to transform industrial operations — supply chain, production, maintenance, quality.

Our portfolio includes over ten active startups (Oplit, Flowlity, Mercateam, Fabriq, Kraaft, Relief, Bonx…), at different stages — from inception to scale-up. They share a common DNA: concrete operational problems, demanding industrial clients, and a strong need for profiles who can bridge the gap between technology and the field.

This role is not tied to a specific startup at the time you read this. It's an open call for FDE profiles we want to know before a need becomes urgent. When a portfolio startup opens this type of role, we introduce our validated pool candidates first.

What is a Forward Deployed Engineer?

The FDE is a profile born at Palantir and now a benchmark in deployed AI. It's neither a consultant nor a pure engineer. It's someone who codes as much as they listen, understands a business problem as well as a data architecture, and only considers a job done when it's running in production — at the client site, in real conditions.

In our portfolio startups, the FDE is often the first technical point of contact with clients. They deploy, adapt, fix, convince, and bring back the insights that shape the product. It's a high-impact role with direct visibility on product decisions — and in some cases, a genuine co-founder-level technical position.

Two very different contexts — one mindset

Early stage — first hire

Some portfolio startups have just come out of the OSS program. They have a product in validation, little or no team, and pilot clients to turn into references. Joining at this stage means being the first technical employee: you build as much as you deploy, you define implementation methods, and you have direct influence on the product roadmap from day one. High ownership, high freedom — and the risks that come with it.

Later stage — scale and structure

Other portfolio startups are already growing, with dozens of clients and an established team. The FDE joins a functioning organisation to take it to the next level: standardising deployments, reducing time-to-value, training client teams, and identifying patterns that allow scaling without growing headcount linearly. High leverage on execution.

What you'll do

Main responsibilities

• Own client deployments end to end — from discovery to production, including configuration, integrations, and testing in real conditions.

• Diagnose clients' operational problems and design the right technical solution — not sell a standard product, but actually solve the problem.

• Be the critical link between product/engineering teams and the field — translate client constraints into actionable specs, and make sure nothing gets built in a vacuum.

• Drive adoption: client team training, operational documentation, post-go-live follow-up. The deployment doesn't end on launch day.

• Actively contribute to the product roadmap — your field insights are information no one else on the team can provide.

Day to day

• 2 to 3 days per week on-site at client locations — factory floors, warehouses, meeting rooms with ops managers and CIOs who have no time to waste.

• Debugging live data pipelines, API integrations, models that behave unexpectedly in production.

• Writing code, scripts, notebooks — to make things work, not to look good.

• Running working sessions with non-technical teams to understand how things actually work, not how the documentation says they do.

What success looks like

At 30 days

You've been through at least one full client deployment. You understand the stack and can onboard a client solo on the core modules. You've already identified a field friction point that nobody had documented.

At 90 days

You're managing 2 to 3 clients independently. Clients ask for "their" FDE on every project. You've contributed to at least one product change based on your field feedback.

At 6 months

You're the internal reference for what actually works with clients. Clients open new projects beyond the initial scope. You're defining or improving the startup's deployment methodology.

Your profile

Must-haves

• 5+ years of experience in a technical role with a strong field component — solutions engineering, implementation, technical consulting, or client-facing engineering.

• You can code — Python, SQL, REST APIs are everyday tools. You can debug a production issue at 6pm at a client site, without a safety net.

• You've managed deployments in complex operational or industrial environments — incomplete data, legacy systems, strict IT constraints, sceptical stakeholders.

• You can read an operational process and identify where technology creates value — without needing the client to spell it out twice.

• You communicate with non-technical people to co-build solutions, not to condescend. You adapt your register without losing precision.

• You're fluent in both French and English — clients are in France, technical references are in English.
  

Technical skills

• Data engineering: structured and semi-structured data manipulation, ETL, advanced SQL, Python (pandas, numpy, automation scripts).

• Integrations: REST APIs, webhooks, ERP/MES/WMS connectors — you can read an API doc and make two systems that don't know each other talk.

• Production environments: cloud deployment (AWS, GCP or Azure), Docker, basic CI/CD. You're not DevOps, but you won't block on a deployment.

• Debugging and diagnostics: able to isolate a problem in a complex stack — logs, monitoring, reproducibility. You don't say 'it works on my machine'.

• Data visualisation and reporting: you can build a dashboard that answers a real business question, not just display numbers.

Nice-to-haves

• Experience with ML / LLM tools in production — fine-tuning, prompt engineering, RAG, model evaluation in real conditions.

• Knowledge of the industrial sector: supply chain, production, or maintenance.

• Background in the Palantir, Databricks, or DataRobot ecosystem — you know the patterns of operational AI deployment.

• Experience building an implementation methodology or deployment playbook from scratch.

This role is probably not for you if…

• You're looking for a 100% technical position with no regular client interaction — client relationships are a core skill here, not a side task.

• You're not comfortable with ambiguity: incomplete specs, messy data, stakeholders who don't know exactly what they want.

• You're not willing to push back on a solution you've been asked to deploy if you think it doesn't match the real need.

• You need a structured organisation with established processes — especially in early stage, you'll be building the processes yourself.

• You need 6 months to be fully operational — here, the first client is waiting by week three.