Software Developer (Research)

• Expand our event-driven strategy across exchange feeds, market news, social media, and infrastructure telemetry.
• Spin up rapid analyses in Python/Jupyter to assess signal quality, and turn what you find into system improvements.

• Build and operate agentic harnesses — tools, context management, evals, guardrails — that do real work against our data and infrastructure, and own their quality in production.
• Apply LLMs where they genuinely win (extraction, classification, triage, research acceleration) and make the call to not use one where they don't.

• Own full-stack performance from the network edge to in-memory stores — instrumenting, monitoring, and debugging production systems alongside operations, keeping SLOs razor-sharp.
• Lead green-field initiatives, design reviews, and post-mortems.

What You Bring

• 3–5 years building real-time or data-intensive systems (we weigh trajectory and depth over the exact number).
• Depth with LLMs, not just usage — how these models actually behave (context and its failure modes, tool use, structured output, cost/latency trade-offs, bounding hallucination) and how to build the harness around one: tools, memory, retries, evals, human-in-the-loop boundaries. Bring something real you shipped and can walk us through end to end, including what broke.
• Strong engineering fundamentals in Python, Go, or Rust. A preference, not a gate — deep systems experience in another serious language transfers.
• Deep knowledge of network programming and protocols — TCP/UDP/IP, DNS, BGP, HTTP(S), WebSocket, QUIC.
• Rapid POCs to production — validate statistical significance, iterate quickly, ship, and explain the result to technical and non-technical audiences alike.
• Bachelor's or Master's in Computer Science, Data Science, Mathematics, or a related discipline.

This is a competitive, high-stakes environment, and we're looking for people who want that
rather than tolerate it. You'll likely recognize yourself here if:

• You want problems that are genuinely hard and measurable — where the scoreboard is real and public inside the team.
• You move fast without being sloppy, and you'd rather ship, measure, and correct than deliberate indefinitely.
• You're comfortable being wrong quickly and in public, and you update rather than defend.

• Depth in the current LLM ecosystem — agent frameworks, MCP, RAG architectures, eval tooling, or local/self-hosted inference.
• Demonstrated success squeezing latency from NGINX, Envoy, or custom load-balancing layers.
• Background in high-pressure domains — trading floors, esports, competitive athletics, or hackathons.
• Open-source contributions, technical blogging, or conference speaking.

• Mentorship with experienced software developers, database administrators, and technical project managers
• Continuous learning through paid postgraduate degrees, Dev Lightning talks, online learning support and 1 on 1 language tutoring with Berlitz
• 40 hours of paid volunteer work at the organization of your choice
• Bi-weekly social activities, monthly wellness plan, on-site weekly massages, and games room
• Enjoy daily catered meals (breakfast and lunch) with unlimited snacks and beverages
• Competitive salary, matching RRSP

Find out more about all of our perks & benefits at https://drw.com/work-at-drw/montreal-benefits.