quality_assurance · security · workflow

Leidos uses Sourcegraph Cody to accelerate software development and cut senior developer mentoring time

Leidos needed an AI coding assistant that met the highest security standards for government and defense clients, avoided LLM lock-in as the field rapidly evolved, and could retrieve context from entire code repositories rather than just the currently open editor tab.

How it works
Common implementation structure
How this type of workflow is generally built, generalized across documented cases — not tied to any one vendor's stack. Click any stage to read what happens there. Specific products that implement these stages appear in “Tools commonly seen” below.
Stage 1 · Engineer invokes Cody
Engineers use Cody every day for all coding tasks.
Tools used
CodySourcegraph
Outcome

Cody accelerated Oracle-to-PostgreSQL migrations to 80% to 90% complete within minutes, and senior developers reduced time guiding junior developers from about eight hours of the week to two. Engineers also save time writing documentation, generate boilerplate in seconds, and debug significantly faster.

What failed first

Many evaluated alternative coding assistants were eliminated for insufficient security and privacy controls, and their context retrieval was limited to only the open editor tab rather than full repositories.

Results
Time savedeight hours of the week... cut this down to two
Volume80% to 90%
Source

https://sourcegraph.com/case-studies/cody-leidos-maximizing-efficiency-heightened-security-ai-race

How we source this →

Grounding & classification
Source type: vendor customer story
23 fields verified against source quotes.
agent assistcode generationragcode diff prmetric backednamed customerproduction runtime claimedtools describedworkflow describedprofessional servicespublic safety defensecycle time reductionemployee productivitytime savedvendor customer storyquality assuranceagentic task execution