We Let 40 Engineers Loose With Coding Agents. The Bill Was Brutal.

A team of 40 engineers using Claude Code coding agents saw a 340% increase in AI costs, reaching $20K/month in unexpected spend due to raw API keys without per-developer budgets or team caps. The team implemented LiteLLM as a proxy to enforce per-developer budgets, team caps, and model restrictions, cutting costs and providing visibility. The solution allows individual caps of $100/month and team budgets of $2,000/month, with alerts at 50% consumption.

Your engineering team adopted Claude Code last month. Productivity went up. Then the bill came in. 340% increase. Nobody budgeted for this. Nobody even knew who spent what. A single coding agent session makes 50-200 API calls. Claude Sonnet 4 processes 100K+ context windows on every call. One developer running sessions all day burns $50-100. Scale that to 40 engineers and you hit $20K/month in unexpected AI spend. The root cause: raw API keys. No per-developer budgets. No team caps. No visibility. You find out about the problem when the invoice arrives. We put LiteLLM https://github.com/BerriAI/litellm between our coding agents and the LLM providers. Every call flows through the proxy, gets tracked, gets budget-checked. Took maybe 15 minutes to set up. Each engineer gets a virtual key with a hard budget cap: curl -X POST 'http://litellm-proxy:4000/key/generate' \ -H 'Authorization: Bearer sk-master' \ -d '{ "key alias": "alice-claude-code", "max budget": 100, "budget duration": "1mo", "models": "claude-sonnet-4-20250514", "gpt-4.1-mini" , "tpm limit": 1000000, "rpm limit": 100 }' $100/month cap. Auto-resets. Rate-limited so a runaway loop can't burn through it in 10 minutes. The developer just changes one env var: export ANTHROPIC BASE URL=http://litellm-proxy:4000 export ANTHROPIC API KEY=sk-alice-generated-key Claude Code doesn't know it's going through a gateway. No SDK changes, no config files, nothing. Individual caps are good. Team budgets catch the case where 20 developers each spending $90 still adds up to $1,800: curl -X POST 'http://litellm-proxy:4000/team/new' \ -H 'Authorization: Bearer sk-master' \ -d '{ "team alias": "backend-eng", "max budget": 2000, "budget duration": "1mo", "models": "claude-sonnet-4-20250514", "gpt-4.1-mini", "gpt-4.1" }' Budget checks happen at every level: key, team, org. If any limit is hit, the request gets rejected with a clear error. No silent failures. Not every task needs Claude Opus $15/M input tokens . Most coding agent work, autocomplete, test generation, docs, that's Sonnet 4 $3/M or GPT-4.1-mini $0.40/M territory. We give junior devs access to cost-effective models only. Senior engineers get the full menu. If an intern's agent tries to call Opus, the request is rejected before any tokens are consumed. This is the part that actually made our CFO happy: response = client.chat.completions.create model="claude-sonnet-4", messages= {"role": "user", "content": "Refactor this function..."} , extra body={ "metadata": { "tags": "project:payments-refactor", "team:backend", "agent:claude-code" } } Now instead of "AI costs $50K/month" the conversation becomes "the payments team spent $12K on Claude Sonnet for their Q3 refactor, saving 3 weeks of engineering time." Without controls, Month 3 of org-wide agent rollout looks like this: With per-developer caps $100/mo and team budgets $2,000/mo , you cap exposure at a number you actually chose. Alerts fire at 50% consumption, giving you 2 weeks to adjust. LangSmith launched their LLM Gateway https://langchain.com/blog/introducing-llm-gateway recently. Fair comparison: For coding agents processing proprietary source code, the self-hosted part matters a lot. Start proxy litellm --config config.yaml Create team curl -X POST 'http://localhost:4000/team/new' \ -H 'Authorization: Bearer sk-master' \ -d '{"team alias": "engineering", "max budget": 5000, "budget duration": "1mo"}' Generate developer key curl -X POST 'http://localhost:4000/key/generate' \ -H 'Authorization: Bearer sk-master' \ -d '{"team id": "TEAM ID", "key alias": "dev-key", "max budget": 100, "budget duration": "1mo"}' Developer sets env var export ANTHROPIC BASE URL=http://litellm-proxy:4000 export ANTHROPIC API KEY=sk-generated-key 15 minutes. Every coding agent call gets tracked, budget-checked, and attributed. Full walkthrough with screenshots: docs.litellm.ai/blog/coding-agent-spend-control https://docs.litellm.ai/blog/coding-agent-spend-control Ran into similar agent cost problems? Curious what approaches other teams are using.